WO2013121776A1

WO2013121776A1 - A marching memory, a bidirectional marching memory, a complex marching memory and a computer system, without the memory bottleneck

Info

Publication number: WO2013121776A1
Application number: PCT/JP2013/000760
Authority: WO
Inventors: Tadao Nakamura; Michael J. Flynn
Original assignee: Tadao Nakamura; Flynn Michael J
Priority date: 2012-02-13
Filing date: 2013-02-13
Publication date: 2013-08-22
Also published as: CN104040635A; JP2015510216A; CN107093463A; CN104040635B; JP6093379B2; TW201737251A; JP6218294B2; US20140344544A1; US11164612B2; TWI607454B; KR20140102745A; EP2815403B1; EP2815403A4; EP3477645A1; EP3982366A1; CN107093463B; KR101689939B1; TWI637396B; KR20160116040A; US10573359B2

Abstract

A marching memory includes an array of memory units (U₁, U₂, U₃,........., U_n-1, U_n), each of the memory units having a sequence of bit-level cells (M₁₁, M₂₁, M₃₁,........, M_m-1,1, M_m1) so as to store information of byte size or word size. Each of the bit-level cells encompasses a transfer-transistor (Q₁₁₁) having a first main-electrode connected to a clock signal supply line (CLOCK) through a first delay element (D₁₁₁) and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array, through a second delay element (D₁₁₂), a reset-transistor (Q₁₁₂) having a control-electrode connected to the clock signal supply line, and a capacitor(C₁₁) connected in parallel with the reset-transistor.

Description

A MARCHING MEMORY, A BIDIRECTIONAL MARCHING MEMORY, A COMPLEX MARCHING MEMORY AND A COMPUTER SYSTEM, WITHOUT THE MEMORY BOTTLENECK

The instant invention relates to new memories and new computer systems using the new memories, which operate at a low energy consumption and high speed.

Since von Neumann and others more than 60 years ago developed a stored program electronic computer, the fundamental memory accessing principle has not been changed. While the processing speeds of computers have increased significantly over the years for whole range of high performance computing (HPC) applications, it has been accomplished either by device technology or by schemes that avoid memory accessing (such as using cache). However, the memory accessing time still limits performance. Currently computer systems use many processors 11 and many large-scale main memories 331, as illustrated in Fig.1.

The computer system illustrated in Fig. 1 includes a processor 11, a cache memory (321a, 321b) and a main memory 331. The processor 11 includes a control unit 111 having a clock generator 113 configured to generate a clock signal, an arithmetic logic unit (ALU) 112 configured to execute arithmetic and logic operations synchronized with the clock signal, a instruction register file (RF) 322a connected to the control unit 111 and a data register file (RF) 322b connected to the ALU 112. The cache memory (321a, 321b) has an instruction cache memory 321a and a data cache memory321b. A portion of the main memory 331 and the instruction cache memory 321a are electrically connected by wires and/or buses, which limits the memory access time (or having the Von Neumann bottleneck).351. The remaining portion of the main memory 331, and the data cache memory 321b are electrically connected to enable a similar memory access.351. Furthermore, wires and/or buses, which implement memory access.352, electrically connect between the data cache memory 321b and the instruction cache memory 321a, and the instruction register file 322a and the data register file 322b.

Even though the HPC systems are expected to operate at high speed and low energy consumption, there are speed limitations due to the

memory accessing bottlenecks

351, 352. The

bottlenecks

351, 352 are ascribable to the wirings between processors 11 and the main memory 331, because the wire length delays access to the computers and stray capacitance existing between wires cause additional delay. Such capacitance requires more power consumption that is proportional to the processor clock frequency in 11.

Currently some HPC processors are implemented using several vector arithmetic pipelines. This vector processor makes better use of memory bandwidth and is a superior machine for HPC applications that can be expressed in vector notation. The vector instructions are made from loops in a source program and each of these vector instructions is executed in an arithmetic pipeline in a vector processor or corresponding units in a parallel processor. The results of these processing schemes give the same results.
However, even the vector processor based system has the

memory bottleneck

351, 352 between all the units. Even in a single system with a wide memory and large bandwidth, the

same bottleneck

351, 352 appears and if the system consists of many of the same units as in a parallel processor, and the

bottleneck

351, 352 is unavoidable.

There are two essential memory access problems in the conventional computer systems. The first problem is wiring lying not only between memory chips and caches or between these two units even on a chip but also inside memory systems. Between chips the wiring between these two chips/units results in more dynamic power consumption due to capacity and the wire signal time delay. This is extended to the internal wire problems within a memory chip related to access lines and the remaining read/write lines. Thus in both inter and intra wiring of memory chips, there exists energy consumption caused by the capacitors with these wires.

The second problem is the

memory bottleneck

351, 352 between processor chip, cache and memory chips. Since the ALU can access any part of cache or memory, the

access path

351, 352 consists of global wires of long length. These paths are also limited in the number of wires available. Such a bottleneck seems to be due to hardware such as busses. Especially when there is a high speed CPU and a large capacity of memory, the apparent bottleneck is basically between these two.

The key to removing the bottleneck is to have the same memory clock cycle as the CPU's. First, addressing proceeding must be created to improve memory access. Secondly the time delay due to longer wires must be significantly reduced both inside memory and outside memory.

By solving these two issues, a fast direct coupling between memory and the CPU is made, which fact enables a computer without the Memory Bottleneck. The processor and periphery of the processor consumes 70% of the total energy because of these problems, which is divided into 42 percent for supplying instructions and 28 percent for data as illustrated in Fig. 53. The wiring problems generate not only power consumption but also time delay of signals. Overcoming the wiring problems implies the elimination of

bottlenecks

351, 352 that limits the flow of data/instructions. If we could remove the wirings of intra/inter chips, the problems of power consumption, time delay and

memory bottlenecks

351, 352 would be solved.

An aspect of the present invention inheres in a marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising: (a) a transfer-transistor having a first main-electrode connected to a clock signal supply line through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of the memory units, through a second delay element; (b) a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and (c) a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor, wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of the memory units.

Here, the first main-electrode shall be assigned as a source electrode or a drain electrode for a field effect transistor (FET), a static induction transistor (SIT), a high electron mobility transistor (HEMT), and others, and the second main-electrode is the drain electrode if the first main-electrode is assigned as the source electrode. Alternatively, the second main-electrode is the source electrode if the first main-electrode is assigned as the drain electrode for FET, SIT, and HEMT etc. Similarly, the first main-electrode shall be assigned as an emitter electrode or a collector electrode for a bipolar junction transistor (BJT), and the second main-electrode is the collector electrode if the first main-electrode is assigned as the emitter electrode. Alternatively, the second main-electrode is the emitter electrode if the first main-electrode is assigned as the collector electrode for BJT. And, the control-electrode is a gate electrode for FET, SIT, and HEMT, etc., and a base electrode for BJT.

Another aspect of the present invention inheres in a bidirectional-marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising: (a) a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element; (b) a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential; (c) a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element; (d) a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential; (e) a forward capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor; and (f) a backward capacitorconfigured to store the information of the bit-level cell, connected in parallel with the backward reset-transistor, wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the forward capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the backward capacitor to the first neighboring bit-level cell.

Still another aspect of the present invention inheres in a bidirectional-marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising: (a) a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element; (b) a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential; (c) a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element; (d) a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential; and (e) a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor, wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.

Still another aspect of the present invention inheres in a complex marching memory encompassing a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size. Here, wherein each of the memory units transfers synchronously with a clock signal synchronized with the CPU's clock signal, step by step, toward an output side of corresponding marching memory block from an input side of the corresponding marching memory block, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.

Still another aspect of the present invention inheres in a complex marching memory encompassing a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size. Here, each of the memory units transfers synchronously with a first clock signal, step by step, toward a first edge side of corresponding marching memory block from a second edge side of the corresponding marching memory block opposing to the first edge side, and further, each of the memory units transfers synchronously with a second clock signal, step by step, toward the second edge side from the first edge side, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.

Still another aspect of the present invention inheres in a computer system comprising a processor and a marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the marching main memory to the processor, the marching main memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising: (a) a transfer-transistor having a first main-electrode connected to a clock signal supply line through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of the memory units, through a second delay element; (b) a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and (c) a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor, wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of the memory units.

Still another aspect of the present invention inheres in a computer system comprising a processor and a bidirectional marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the bidirectional marching main memory to the processor, the bidirectional marching main memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising: (a) a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element; (b) a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential; (c) a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element; (d) a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential; and (e) a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor, wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.

Still another aspect of the present invention inheres in a computer system encompassing a processor and a marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the marching main memory to the processor, the marching main memory encompassing a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size. Here, each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.

Still another aspect of the present invention inheres in a computer system encompassing a processor and a bidirectional marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the bidirectional marching main memory to the processor, the bidirectional marching main memory encompassing a plurality of bidirectional marching memory blocks being deployed spatially, each of the bidirectional marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size. Here, each of the memory units transfers synchronously with a first clock signal, step by step, toward a first edge side of corresponding marching memory block from a second edge side of the corresponding marching memory block opposing to the first edge side, and further, each of the memory units transfers synchronously with a second clock signal, step by step, toward the second edge side from the first edge side, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.

Fig. 1 illustrates a schematic block diagram illustrating an organization of a conventional computer system; Fig. 2 illustrates a schematic block diagram illustrating a fundamental organization of a computer system pertaining to a first embodiment of the present invention; Fig. 3 illustrates an array of memory units implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention, and a transfer of information in the marching main memory; Fig. 4 illustrates an example of a transistor-level representation of the cell-array in the marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 5 illustrates an enlarged transistor-level representation of the cell-array in the marching main memory used in the computer system pertaining to the first embodiment of the present invention, focusing to four neighboring bit-level cells; Fig. 6 illustrates a further enlarged transistor-level representation of a single bit-level cell in the marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 7A illustrates a schematic example of the response of the transistor to the waveform of a clock signal configured to be applied to the marching main memory used in the computer system pertaining to the first embodiment of the present invention, illustrating a case when a signal "1" is transferred from the previous stage; Fig. 7B illustrates another schematic example of the response of the transistor to the waveform of the clock signal configured to be applied to the marching main memory used in the computer system pertaining to the first embodiment of the present invention, illustrating another case when a signal "0" is transferred from the previous stage; Fig. 7C illustrates an actual example of the responses of the transistors to the waveform of a clock signal configured to be applied to the marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 8 illustrates a detailed example of a bit-level cell used in the marching main memory for the computer system pertaining to the first embodiment of the present invention; Fig. 9 illustrates an example of actual plan view implementing the bit-level cell illustrated in Fig. 8; Fig. 10 illustrates a cross-sectional view taken on line A-A in the plan view illustrated in Fig. 9; Fig. 11 illustrates another enlarged transistor-level representation of the single bit-level cell in combination with an inter-unit cell adapted for a marching main memory used in the computer system pertaining to a modification of the first embodiment of the present invention; Fig. 12 illustrates an example of actual plan view implementing the bit-level cell illustrated in Fig. 11; Fig. 13 illustrates an enlarged transistor-level representation of the cell-array in combination with corresponding inter-unit cells, in the marching main memory used in the computer system pertaining to the modification of the first embodiment of the present invention, focusing to two neighboring bit-level cells; Fig. 14(a) illustrates a timing diagram of a response of the bit-level cell illustrated in Fig. 13, and Fig. 14(b) illustrates a next timing diagram of a next response of the next bit-level cell illustrated in Fig. 13, to a waveform of a clock signal. Fig. 15 illustrates an actual example of the responses of the transistors to the waveform of a clock signal configured to be applied to the marching main memory used in the computer system pertaining to the modification of the first embodiment of the present invention; Figs. 16 (a)-(d) illustrate four modes of signal-transferring operations, respectively, focusing to the bit-level cell illustrated in Figs. 11 and 13, in the marching main memory used in the computer system pertaining to the modification of the first embodiment of the present invention; Fig. 17 illustrates a still another enlarged transistor-level representation of the single bit-level cell in combination with an inter-unit cell adapted for a marching main memory used in the computer system pertaining to another modification (second modification) of the first embodiment of the present invention; Fig. 18 illustrates an enlarged transistor-level representation of the cell-array in combination with corresponding inter-unit cells, in the marching main memory used in the computer system pertaining to the second modification of the first embodiment of the present invention, focusing to two neighboring bit-level cells; Fig. 19 illustrates a yet still another enlarged transistor-level representation of the single bit-level cell in combination with an inter-unit cell adapted for a marching main memory used in the computer system pertaining to a still another modification (third modification) of the first embodiment of the present invention; Fig. 20 illustrates an enlarged transistor-level representation of the cell-array in combination with corresponding inter-unit cells, in the marching main memory used in the computer system pertaining to the third modification of the first embodiment of the present invention, focusing to two neighboring bit-level cells; Fig. 21 illustrates an actual example of the responses of the transistors to the waveform of a clock signal configured to be applied to the marching main memory used in the computer system pertaining to the third modification of the first embodiment of the present invention; Figs. 22 (a)-(d) illustrate four modes of signal-transferring operations, respectively, focusing to the bit-level cell illustrated in Figs. 20 and 21, in the marching main memory used in the computer system pertaining to the third modification of the first embodiment of the present invention; Fig. 23 illustrates a gate-level representation of the cell-array illustrated in Fig. 4; Fig. 24 illustrates an array of memory units implementing a reverse directional marching main memory used in the computer system pertaining to the first embodiment of the present invention, and a reverse directional transfer of information in the reverse directional marching main memory; Fig. 25(a) illustrates an example of a transistor-level circuit configuration of cell array implementing i-th row of the reverse directional marching main memory illustrated in Fig. 24, and Fig. 25(b) illustrates an example of the response of the transistor to the waveform of a clock signal configured to be applied to the reverse directional marching main memory illustrated in Fig. 24; Fig. 26 illustrates a gate-level representation of the cell-array implementing i-th row in the reverse directional marching main memory illustrated in Fig. 25 (a); Fig. 27 illustrates a time-domain relationship between the memory unit streaming time in a marching main memory and the clock cycle in a processor (CPU) in the computer system pertaining to the first embodiment of the present invention; Fig. 28 illustrates schematically an organization of the computer system pertaining to the first embodiment of the present invention, in which the memory bottleneck is disappeared between the processor (CPU) and the marching memory structure including the marching main memory, in the computer system pertaining to the first embodiment of the present invention; Fig. 29(a) illustrates a forward data stream flowing from the marching memory structure, which includes the marching main memory, to the processor (CPU) and backward data stream flowing from the processor (CPU) to the marching memory structure in the computer system pertaining to the first embodiment of the present invention, and Fig. 29(b) illustrates bandwidths established between the marching memory structure and the processor (CPU) under an ideal condition that the memory unit streaming time of the marching memory structure is equal to the clock cycle of the processor (CPU); Fig. 30(a) schematically illustrates an extremely high-speed magnetic tape system, comparing with the computer system illustrated in Fig. 30(b), which corresponds to the computer system pertaining to the first embodiment of the present invention; Fig. 31 (a) illustrates a concrete image of a marching behavior of information (a forward marching behavior), in which information marches (shifts) side by side toward right-hand direction in a one-dimensional marching main memory, Fig. 31 (b) illustrates a staying state of the one-dimensional marching main memory, and Fig. 31 (c) illustrates a concrete image of a reverse-marching behavior of information (a backward marching behavior), in which information marches (shifts) side by side toward left-hand direction in the one-dimensional marching main memory in the computer system pertaining to the first embodiment of the present invention; Fig. 32 illustrates an example of a transistor-level circuit configuration of the one-dimensional marching main memory, which can achieve the bidirectional transferring behavior illustrated in Figs. 31 (a)-(c), configured to store and transfer bi-directionally instructions or scalar data in the computer system pertaining to the first embodiment of the present invention; Fig. 33 illustrates another example of a transistor-level circuit configuration of the one-dimensional marching main memory, incorporating isolation transistors between memory units, which can achieve the bidirectional transferring behavior illustrated in Figs. 31 (a)-(c), configured to store and transfer bi-directionally instructions or scalar data in the computer system pertaining to the first embodiment of the present invention; Fig. 34 illustrates a generic representation of the gate-level circuit configuration of the one-dimensional marching main memory illustrated in Fig. 32; Fig. 35(a) illustrates a bidirectional transferring mode of instructions in a one-dimensional marching main memory adjacent to a processor, the instructions moves toward the processor, and moves from / to the next memory arranged at left-hand side, Fig. 35(b) illustrates a bidirectional transferring mode of scalar data in a one-dimensional marching main memory adjacent to an ALU, the scalar data moves toward the ALU and moves from / to the next memory, and Fig. 35(c) illustrates a uni-directional transferring mode of vector/streaming data in a one-dimensional marching main memory adjacent to a pipeline, the vector/streaming data moves toward the pipeline, and moves from the next memory; Fig. 36(a) compares with Fig. 36(b), illustrating an inner configuration of existing memory, in which each memory unit is labeled by an address, and Fig. 36(b) illustrates an inner configuration of present one-dimensional marching main memory, in which the positioning of individual memory unit is at least necessary to identify the starting point and ending point of a set of successive memory units in vector/streaming data. Fig. 37(a) illustrates an inner configuration of present one-dimensional marching main memory, in which the positioning of individual memory unit is at least necessary to identify the starting point and ending point of a set of successive memory units in vector instruction, Fig. 37(b) illustrates an inner configuration of present one-dimensional marching main memory for scalar data.. Fig. 37(c) illustrates an inner configuration of present one-dimensional marching main memory, in which position indexes are at least necessary to identify the starting point and ending point of a set of successive memory units in vector/streaming data; Fig. 38(a) illustrates schematically an example of an overall configuration of present marching main memory implemented by a plurality of pages for vector/streaming data case, Fig. 38(b) illustrates schematically an example of a configuration of one of the pages, each of the page is implemented by a plurality of files for vector/streaming data case, and Fig. 38(c) illustrates schematically an example of a configuration of one of the files, each of the file is implemented by a plurality of memory units for vector/streaming data case, in the computer system pertaining to the first embodiment of the present invention; Fig. 39(a) illustrates schematically an example of an overall configuration of present marching main memory implemented by a plurality of pages for programs/scalar data case, where each pages has its own position index as address, Fig. 39(b) illustrates schematically an example of a configuration of one of the pages and the driving positions of the page, using digits in the binary system, each of the page is implemented by a plurality of files for programs/scalar data case, and each file has its own position index as address, and Fig. 39(c) illustrates schematically an example of a configuration of one of the files and the driving positions of the file, using digits in the binary system, each of the file is implemented by a plurality of memory units for programs/scalar data case, where each memory units has its own position index as address, in the computer system pertaining to the first embodiment of the present invention; Fig. 40(a) illustrates schematically the speed/capability of the existing memory compared with that of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, and Fig. 40(b) illustrates schematically the speed/capability of the marching main memory compared with that of the existing memory illustrated in Fig. 40(a); Fig. 41(a) illustrates schematically the speed/capability of the worst case of the existing memory for scalar instructions compared with that of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, and Fig. 41(b) illustrates schematically the speed/capability of the marching main memory compared with that of the worst case of the existing memory illustrated in Fig. 41(a); Fig. 42(a) illustrates schematically the speed/capability of the typical case of the existing memory for scalar instructions compared with that of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, and Fig. 42(b) illustrates schematically the speed/capability of the marching main memory compared with that of the typical case of the existing memory illustrated in Fig. 42(a); Fig. 43(a) illustrates schematically the speed/capability of the typical case of the existing memory for scalar data case compared with that of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, and Fig. 43(b) illustrates schematically the speed/capability of the marching main memory compared with that of the existing memory illustrated in Fig. 43(a); Fig. 44(a) illustrates schematically the speed/capability of the best case of the existing memory for streaming data and data parallel case compared with that of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, and Fig. 44(b) illustrates schematically the speed/capability of the marching main memory compared with that of the best case of the existing memory illustrated in Fig. 44(a); Fig. 45 illustrates an example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 46 illustrates another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 47 illustrates a still another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 48 illustrates a yet still another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 49 illustrates a further another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 50 illustrates a further another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 51 illustrates a further another example of the array of two-dimensional memory units, each of the memory units storing and transferring data or instructions, implementing a marching main memory used in the computer system pertaining to the first embodiment of the present invention; Fig. 52(a) illustrates device level energy consumption in current microprocessors, decomposing into static and dynamic energy consumptions, Fig. 52(b) illustrates net and overhead of the power consumption in the dynamic energy consumption illustrated in Fig. 52(a), and Fig. 52(c) illustrates the net energy consumption in the current microprocessors; Fig. 53 illustrates an actual energy consumption distribution over a processor including registers and caches in the conventional architecture, estimated by Dally; Fig. 54(a) illustrates energy consumption in the conventional cache-based architecture, decomposing the energy consumption in the cache memory into static and dynamic energy consumptions, and Fig. 54(b) illustrates energy consumption in the computer system according to a third embodiment of the present invention, decomposing the energy consumption in the marching cache memory into static and dynamic energy consumption. Fig. 55 illustrates a schematic block diagram illustrating an organization of a computer system pertaining to a second embodiment of the present invention; Fig. 56 illustrates a schematic block diagram illustrating an organization of a computer system pertaining to a third embodiment of the present invention; Fig. 57(a) illustrates a combination of arithmetic pipelines and marching register units in the computer system pertaining to the third embodiment of the present invention, and Fig. 57(b) illustrates an array of marching cache units in the computer system pertaining to the third embodiment of the present invention; Fig. 58 illustrates a schematic block diagram of an organization of a computer system implemented by a combination of a single processor core, a marching-cache memory and a marching-register file in accordance with a modification of the third embodiment of the present invention; Fig. 59 illustrates a schematic block diagram of an organization of a computer system implemented by a combination of a single arithmetic pipeline, a marching-cache memory and a marching-vector register file in accordance with another modification of the third embodiment of the present invention; Fig. 60 illustrates a schematic block diagram of an organization of a computer system implemented by a combination of a plurality of processor cores, a marching-cache memory and a marching-register file in accordance with a still another modification of the third embodiment of the present invention; Fig. 61 illustrates a schematic block diagram of an organization of a computer system implemented by a combination of a plurality of arithmetic pipelines, a marching-cache memory and a marching-vector register file in accordance with a yet still another modification of the third embodiment of the present invention; Fig. 62(a) illustrates a schematic block diagram of an organization of a conventional computer system implemented by a combination of a plurality of arithmetic pipelines, a plurality of conventional cache memories, a plurality of conventional-vector register files (RFs) and a conventional main memory, in which bottleneck is established between the conventional cache memories and the conventional main memory, and Fig. 62(b) illustrates a schematic block diagram of an organization of a computer system implemented by a combination of a plurality of arithmetic pipelines, a plurality of marching cache memories, a plurality of marching-vector register files and a marching main memory, in which no bottleneck is established, in accordance with a yet still another modification of the third embodiment of the present invention; Fig. 63 illustrates a schematic block diagram illustrating an organization of a high performance computing (HPC) system pertaining to a fourth embodiment of the present invention; Fig. 64 illustrates a schematic block diagram illustrating an organization of a computer system pertaining to a fifth embodiment of the present invention; Fig. 65(a) illustrates a cross-sectional view of a three-dimensional marching main memory used in the computer system pertaining to the fifth embodiment of the present invention, Fig. 65(b) illustrates a cross-sectional view of a three-dimensional marching-cache used in the computer system pertaining to the fifth embodiment of the present invention, and Fig. 65(c) illustrates a cross-sectional view of a three-dimensional marching-register file used in the computer system pertaining to the fifth embodiment of the present invention; Fig. 66 illustrates a perspective view of a three-dimensional configuration used in the computer system pertaining to the fifth embodiment of the present invention; Fig. 67 illustrates a perspective view of another three-dimensional configuration used in the computer system pertaining to the fifth embodiment of the present invention; Fig. 68 illustrates a cross-sectional view of the three-dimensional configuration illustrated in Fig. 67; Fig. 69 illustrates a cross-sectional view of another three-dimensional configuration used in the computer system pertaining to the fifth embodiment of the present invention; Fig. 70 illustrates schematically a cross-sectional view of the three-dimensional configuration of a fundamental core of the computer system for executing the control processing, by representing control paths in the computer system pertaining to the fifth embodiment of the present invention; Fig. 71 illustrates schematically a cross-sectional view of the three-dimensional configuration of a fundamental core of the computer system for executing the scalar data processing, by representing data-paths for scalar data in the computer system pertaining to the fifth embodiment of the present invention; Fig. 72 illustrates schematically a cross-sectional view of the three-dimensional configuration of a fundamental core of the computer system for executing the vector/streaming data processing, by representing data-paths for vector/streaming data in the computer system pertaining to the fifth embodiment of the present invention; Fig. 73 illustrates schematically a cross-sectional view of the three-dimensional configuration of a fundamental core of the computer system, configured to execute the scalar data part of the computer system, where a plurality of processing units (CPUs) execute not only scalar data but also vector/streaming data, and pipelined ALUs are included in the processing units, by representing the combination of scalar data-path and the control path for the computer system pertaining to the fifth embodiment of the present invention; Fig. 74 illustrates a bit-level parallel processing of scalar/vector data in MISD architecture; Fig. 75 illustrates a parallel processing of vector data in SIMD architecture; Fig. 76 illustrates a typical chaining in vector processing; Fig. 77 illustrates a parallel processing of scalar/vector data in MISD architecture; Fig. 78 illustrates a parallel processing of scalar/vector data in MISD architecture; Fig. 79(a) illustrates a plan view of a representative conventional DRAM delineated on a single semiconductor chip, and Fig. 79(b) illustrates a corresponding plan view of a schematic inner layout of a complex marching memory, which is delineated on the same single semiconductor chip of the conventional DRAM; Fig. 80 (a) illustrates an outer shape of a single marching memory block, Fig. 80 (b) illustrates a partial plan view of the marching memory block illustrated in Fig. 80 (a), which has one thousand of columns, where the marching memory's access time (cycle time) is defined to a single column, and Fig. 80(c) illustrates the conventional DRAM's memory cycle for writing in or reading out the content of the conventional DRAM's one memory element; and Fig. 81 illustrates a schematic plan view of a complex marching memory module.

Various embodiments of the present invention will be described with reference to the accompanying drawings. It is to be noted that the same or similar reference numerals are applied to the same or similar parts and elements throughout the drawings, and the description of the same or similar parts and elements will be omitted or simplified. Generally and as it is conventional in the representation of semiconductor devices, it will be appreciated that the various drawings are not drawn to scale from one figure to another nor inside a given figure, and in particular that the layer thicknesses are arbitrarily drawn for facilitating the reading of the drawings. In the following description specific details are set forth, such as specific materials, processes and equipment in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known manufacturing materials, processes and equipment are not set forth in detail in order not to unnecessarily obscure the present invention. Prepositions, such as "on", "over", "under", "beneath", and "normal" are defined with respect to a planar surface of the substrate, regardless of the orientation in which the substrate is actually held. A layer is on another layer even if there are intervening layers.

Although nMOS transistors are illustrated as transfer-transistors and reset-transistors in transistor-level representations of bit-level cells in Figs. 4, 5, 6, 8, 11, 13, 16-20, 22, 25 and 32, etc., pMOS transistors can be used as the transfer-transistors and the reset-transistors, if the opposite polarity of the clock signal is employed.

--FIRST EMBODIMENT--
(FUNDAMENTAL ORGANIZATION OF COMPUTER SYSTEM)
As illustrated in Fig. 2, a computer system pertaining to a first embodiment of the present invention encompasses a processor 11 and a marching main memory 31. The processor 11 includes a control unit 111 having a clock generator 113 configured to generate a clock signal, and an arithmetic logic unit (ALU) 112 configured to execute arithmetic and logic operations synchronized with the clock signal. As illustrated in Fig. 3, the marching main memory 31 encompasses an array of memory units U₁, U₂, U₃,........., U_n-1, U_n, each of memory units U₁, U₂, U₃,........., U_n-1, U_n having a unit of information including word size of data or instructions, input terminals of the array and output terminals of the array. As illustrated in Fig. 3, the marching main memory 31 stores the information in each of memory units U₁, U₂, U₃,........., U_n-1, U_n and transfers the information synchronously with the clock signal, step by step, toward the output terminals, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

As illustrated in Fig. 2, the marching main memory 31 and the processor 11 are electrically connected by a plurality of joint members 54. For example, each of joint members 54 may be implemented by a first terminal pin attached to the marching main memory 31, a second terminal pin attached to the processor 11, and an electrical conductive bump interposed between the first and second terminal pins. For the material of the electrical conductive bumps, solder balls, gold (Au) bumps, silver (Ag) bumps, copper (Cu) bumps, nickel-gold (Ni-Au) alloy bumps or nickel-gold-indium (Ni-Au-In) alloy bumps, etc. are acceptable. The resultant data of the processing in the ALU 112 are sent out to the marching main memory 31 through the joint members 54. Therefore, as represented by bidirectional arrow PHI (Greek-letter)₁₂, data are transferred bi-directionally between the marching main memory 31 and the processor 11 through the joint members 54. On the contrary, as represented by uni-directional arrow ETA(Greek-letter)₁₁, as to the instructions movement, there is only one way of instruction-flow from the marching main memory 31 to the processor 11.

As illustrated in Fig. 2, the organization of the computer system pertaining to the first embodiment of the present invention further encompasses an external secondary memory 41 such as disk, an input unit 61, an output unit 62 and input/ output (I/O) interface circuit 63. Similar to a conventional von Neumann computer, the signals or data are received by the input unit 61, and the signals or data are sent from the output unit 62. For instance, known keyboards and known mice can be considered as the input unit 6, while known monitors and printers can be considered as the output unit 62. Known devices for communication between computers, such as modems and network cards, typically serve for both the input unit 61 and the output unit 62. Note that the designation of a device as either the input unit 61 or the output unit 62 depends on the perspective. The input unit 61 takes as input physical movement that the human user provides and converts it into signals that the computer system pertaining to the first embodiment can understand. For example, the input unit 61 converts incoming data and instructions into a pattern of electrical signals in binary code that are comprehensible to the computer system pertaining to the first embodiment, and the output from the input unit 61 is fed to the marching main memory 31 through the I/O interface circuit 63. The output unit 62 takes as input signals that the marching main memory 31 provides through the I/O interface circuit 63. The output unit 62 then converts these signals into representations that human users can see or read, reversing the process of the input unit 61, translating the digitized signals into a form intelligible to the user. The I/O interface circuit 63 is required whenever the processor 11 drives the input unit 61 and the output unit 62. The processor 11 can communicate with the input unit 61 and the output unit 62 through the I/O interface circuit 63. If in the case of different data formatted being exchanged, the I/O interface circuit 63 converts serial data to parallel form and vice-versa. There is provision for generating interrupts and the corresponding type numbers for further processing by the processor 11 if required.

The secondary memory 41 stores data and information on a more long-term basis than the marching main memory 31. While the marching main memory 31 is concerned mainly with storing programs currently executing and data currently being employed, the secondary memory 41 is generally intended for storing anything that needs to be kept even if the computer is switched off or no programs are currently executing. The examples of the secondary memory 41 are known hard disks (or hard drives) and known external media drives (such as CD-ROM drives). These storage methods are most commonly used to store the computer's operating system, the user's collection of software and any other data the user wishes. While the hard drive is used to store data and software on a semi-permanent basis and the external media drives are used to hold other data, this setup varies wildly depending on the different forms of storage available and the convenience of using each. As represented by bidirectional arrow PHI (Greek-letter)₁, data are transferred bi-directionally between the secondary memory 41 and the marching main memory 31 and the processor 11 through existing wire connection 53.

Although the illustration is omitted, in the computer system of the first embodiment illustrated in Fig. 2, the processor 11 may includes a plurality of arithmetic pipelines configured to receive the stored information through the output terminals from the marching main memory 31, and as represented by bidirectional arrow PHI₁₂, data are transferred bi-directionally between the marching main memory 31 and the plurality of arithmetic pipelines through the joint members 54.

In the computer system of the first embodiment illustrated in Fig. 2, there are no buses consisting of the data bus and address bus because the whole computer system has no global wires even in any data exchange between the processor 11 and the marching main memory 31, while the wires or the buses implement the bottleneck in the conventional computer system. There are only short local wires within the marching main memory 31 or connecting portions of the marching main memory 31 with a corresponding ALU 112. As there are no global wires, which generate time delay and stray capacitances between these wires, the computer system of the first embodiment can achieve much higher processing speed and lower power consumption.

(DETAILED CONFIGURATION OF CELL ARRAY IMPLEMENTING THE (MARCHING MAIN MEMORY)
In most conventional computers, the unit of address resolution is either a character (e.g. a byte) or a word. If the unit is a word, then a larger amount of memory can be accessed using an address of a given size. On the other hand, if the unit is a byte, then individual characters can be addressed (i.e. selected during the memory operation). Machine instructions are normally fractions or multiples of the architecture's word size. This is a natural choice since instructions and data usually share the same memory subsystem. Figs. 4 and 5 correspond to transistor-level representations of the cell array implementing the marching main memory 31 illustrated in Fig. 3, and Fig. 23 corresponds to a gate-level representation of the cell array implementing marching main memory 31 illustrated in Fig. 3.

In Fig. 4, the first column of the m * n matrix, which is implemented by a vertical array of cell M₁₁, M₂₁, M₃₁, ........, M_m-1,1, M_m1, represents the first memory unit U₁ illustrated in Fig.3. Here, "m" is an integer determined by word size. Although the choice of a word size is of substantial importance, when computer architecture is designed, word sizes are naturally multiples of eight bits, with 16, 32, and 64 bits being commonly used. Similarly, the second column of the m * n matrix, which is implemented by a vertical array of cell M₁₂, M₂₂, M₃₂, ........, M_m-1,2, M_m2, represents the second memory unit U₂, the third column of the m * n matrix, which is implemented by a vertical array of cell M₁₃, M₂₃, M₃₃, ........, M_m-1,3, M_m3, represents the third memory unit U₃, ........, the (n-1)-th column of the m * n matrix, which is implemented by a vertical array of cell M_1,n-1, M_2,n-1, M_3,n-1, ........, M_m-1,n-1, M_m,n-1, represents the (n-1)-th memory unit U_n-1, and the n-th column of the m * n matrix, which is implemented by a vertical array of cell M_1,n, M_2,n, M_3,n, ........, M_m-1,n, M_m,n, represents the n-th memory unit U_n.

Namely, as illustrated in Fig. 4, the first memory unit U₁of word-size level is implemented by a vertical array of bit-level cell M₁₁, M₂₁, M₃₁, ........, M_m-1,1, M_m1 in the first column of the m * n matrix. The first-column cell M₁₁ on the first row encompasses a first nMOS transistor Q₁₁₁ having a drain electrode connected to a clock signal supply line through a first delay element D₁₁₁ and a gate electrode connected to the output terminal of a first bit-level input terminal through a second delay element D₁₁₂; a second nMOS transistor Q₁₁₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₁₁₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₁₁configured to store the information of the cell M₁₁, connected in parallel with the second nMOS transistor Q₁₁₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₁₁₁ and the drain electrode of the second nMOS transistor Q₁₁₂ serves as an output terminal of the cell M₁₁, configured to deliver the signal stored in the capacitor C₁₁ to the next bit-level cell M₁₂. The first-column cell M₂₁ on the second row encompasses a first nMOS transistor Q₂₁₁ having a drain electrode connected to the clock signal supply line through a first delay element D₂₁₁ and a gate electrode connected to the output terminal of a second bit-level input terminal through a second delay element D₂₁₂; a second nMOS transistor Q₂₁₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₂₁₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₂₁configured to store the information of the cell M₂₁, connected in parallel with the second nMOS transistor Q₂₁₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₂₁₁ and the drain electrode of the second nMOS transistor Q₂₁₂ serves as an output terminal of the cell M₂₁, configured to deliver the signal stored in the capacitor C₂₁ to the next bit-level cell M₂₂. The first-column cell M₃₁ on the third row encompasses a first nMOS transistor Q₃₁₁ having a drain electrode connected to the clock signal supply line through a first delay element D₃₁₁ and a gate electrode connected to the output terminal of a third bit-level input terminal through a second delay element D₃₁₂; a second nMOS transistor Q₃₁₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₃₁₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₃₁configured to store the information of the cell M₃₁, connected in parallel with the second nMOS transistor Q₃₁₂ , wherein an output node connecting the source electrode of the first nMOS transistor Q₃₁₁ and the drain electrode of the second nMOS transistor Q₃₁₂ serves as an output terminal of the cell M₃₁, configured to deliver the signal stored in the capacitor C₃₁ to the next bit-level cell M₃₁. ............ The first-column cell M_(m-1)1 on the (m-1)-th row encompasses a first nMOS transistor Q_(m-1)11 having a drain electrode connected to the clock signal supply line through a first delay element D_(m-1)11 and a gate electrode connected to the output terminal of a (m-1)-th bit-level input terminal through a second delay element D_(m-1)12; a second nMOS transistor Q_(m-1)12 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(m-1)11, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(m-1)1configured to store the information of the cell M_(m-1)1, connected in parallel with the second nMOS transistor Q_(m-1)12 , wherein an output node connecting the source electrode of the first nMOS transistor Q_(m-1)11 and the drain electrode of the second nMOS transistor Q_(m-1)12 serves as an output terminal of the cell M_(m-1)1, configured to deliver the signal stored in the capacitor C_(m-1)1 to the next bit-level cell M_(m-1)12. The first-column cell M_m1 on the m-th row encompasses a first nMOS transistor Q_m11 having a drain electrode connected to the clock signal supply line through a first delay element D_m11 and a gate electrode connected to the output terminal of a m-th bit-level input terminal through a second delay element D_m12; a second nMOS transistor Q_m12 having a drain electrode connected to a source electrode of the first nMOS transistor Q_m11, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_m1configured to store the information of the cell M_m1, connected in parallel with the second nMOS transistor Q_m12 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_m11 and the drain electrode of the second nMOS transistor Q_m12 serves as an output terminal of the cell M_m1, configured to deliver the signal stored in the capacitor C_m1 to the next bit-level cell M_m2.

And, as illustrated in Fig.4, the second memory unit U₂of word-size level is implemented by a vertical array of bit-level cell M₁₂, M₂₂, M₃₂, ........, M_m-1,2, M_m2 in the second column of the m * n matrix. The second column cell M₁₂ on the first row encompasses a first nMOS transistor Q₁₂₁ having a drain electrode connected to the clock signal supply line through a first delay element D₁₂₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₁₁ through a second delay element D₁₂₂; a second nMOS transistor Q₁₂₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₁₂₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₁₂configured to store the information of the cell M₁₂, connected in parallel with the second nMOS transistor Q₁₂₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₁₂₁ and the drain electrode of the second nMOS transistor Q₁₂₂ serves as an output terminal of the cell M₁₂, configured to deliver the signal stored in the capacitor C₁₂ to the next bit-level cell M₁₃. The second column cell M₂₂ on the second row encompasses a first nMOS transistor Q₂₂₁ having a drain electrode connected to the clock signal supply line through a first delay element D₂₂₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₂₁ through a second delay element D₂₂₂; a second nMOS transistor Q₂₂₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₂₂₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₂₂configured to store the information of the cell M₂₂, connected in parallel with the second nMOS transistor Q₂₂₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₂₂₁ and the drain electrode of the second nMOS transistor Q₂₂₂ serves as an output terminal of the cell M₂₂, configured to deliver the signal stored in the capacitor C₂₂ to the next bit-level cell M₂₃. The second column cell M₃₂ on the third row encompasses a first nMOS transistor Q₃₂₁ having a drain electrode connected to the clock signal supply line through a first delay element D₃₂₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₃₁ through a second delay element D₃₂₂; a second nMOS transistor Q₃₂₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₃₂₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₃₂configured to store the information of the cell M₃₂, connected in parallel with the second nMOS transistor Q₃₂₂ , wherein an output node connecting the source electrode of the first nMOS transistor Q₃₂₁ and the drain electrode of the second nMOS transistor Q₃₂₂ serves as an output terminal of the cell M₃₂, configured to deliver the signal stored in the capacitor C₃₂ to the next bit-level cell M₃₃. ............ The second column cell M_(m-1)2 on the (m-1)-th row encompasses a first nMOS transistor Q_(m-1)21 having a drain electrode connected to the clock signal supply line through a first delay element D_(m-1)21 and a gate electrode connected to the output terminal of the previous bit-level cell M_(m-1)1 through a second delay element D_(m-1)22; a second nMOS transistor Q_(m-1)22 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(m-1)21, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(m-1)2configured to store the information of the cell M_(m-1)2, connected in parallel with the second nMOS transistor Q_(m-1)22 , wherein an output node connecting the source electrode of the first nMOS transistor Q_(m-1)21 and the drain electrode of the second nMOS transistor Q_(m-1)22 serves as an output terminal of the cell M_(m-1)2, configured to deliver the signal stored in the capacitor C_(m-1)2 to the next bit-level cell M_(m-1)3. The second column cell M_m2 on the m-th row encompasses a first nMOS transistor Q_m21 having a drain electrode connected to the clock signal supply line through a first delay element D_m21 and a gate electrode connected to the output terminal of the previous bit-level cell M_m1 through a second delay element D_m22; a second nMOS transistor Q_m22 having a drain electrode connected to a source electrode of the first nMOS transistor Q_m21, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_m2configured to store the information of the cell M_m2, connected in parallel with the second nMOS transistor Q_m22 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_m21 and the drain electrode of the second nMOS transistor Q_m22 serves as an output terminal of the cell M_m2, configured to deliver the signal stored in the capacitor C_m2 to the next bit-level cell M_m3.

Furthermore, as illustrated in Fig.4, the third memory unit U₃of word-size level is implemented by a vertical array of bit-level cell M₁₃, M₂₃, M₃₃, ........, M_m-1,3, M_m3 in the third column of the m * n matrix. The third-column cell M₁₃ on the first row encompasses a first nMOS transistor Q₁₃₁ having a drain electrode connected to the clock signal supply line through a first delay element D₁₃₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₁₂ through a second delay element D₁₃₂; a second nMOS transistor Q₁₃₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₁₃₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₁₃configured to store the information of the cell M₁₃, connected in parallel with the second nMOS transistor Q₁₃₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₁₃₁ and the drain electrode of the second nMOS transistor Q₁₃₂ serves as an output terminal of the cell M₁₃, configured to deliver the signal stored in the capacitor C₁₃ to the next bit-level cell. The third-column cell M₂₃ on the second row encompasses a first nMOS transistor Q₂₃₁ having a drain electrode connected to the clock signal supply line through a first delay element D₂₃₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₂₂ through a second delay element D₂₃₂; a second nMOS transistor Q₂₃₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₂₃₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₂₃configured to store the information of the cell M₂₃, connected in parallel with the second nMOS transistor Q₂₃₂ ,wherein an output node connecting the source electrode of the first nMOS transistor Q₂₃₁ and the drain electrode of the second nMOS transistor Q₂₃₂ serves as an output terminal of the cell M₂₃, configured to deliver the signal stored in the capacitor C₂₃ to the next bit-level cell. The third-column cell M₃₃ on the third row encompasses a first nMOS transistor Q₃₃₁ having a drain electrode connected to the clock signal supply line through a first delay element D₃₃₁ and a gate electrode connected to the output terminal of the previous bit-level cell M₃₂ through a second delay element D₃₃₂; a second nMOS transistor Q₃₃₂ having a drain electrode connected to a source electrode of the first nMOS transistor Q₃₃₁, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C₃₃configured to store the information of the cell M₃₃, connected in parallel with the second nMOS transistor Q₃₃₂ , wherein an output node connecting the source electrode of the first nMOS transistor Q₃₃₁ and the drain electrode of the second nMOS transistor Q₃₃₂ serves as an output terminal of the cell M₃₃, configured to deliver the signal stored in the capacitor C₃₃ to the next bit-level cell. ............ The third-column cell M_(m-1)3 on the (m-1)-th row encompasses a first nMOS transistor Q_(m-1)31 having a drain electrode connected to the clock signal supply line through a first delay element D_(m-1)31 and a gate electrode connected to the output terminal of the previous bit-level cell M_(m-1)2 through a second delay element D_(m-1)32; a second nMOS transistor Q_(m-1)32 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(m-1)31, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(m-1)3configured to store the information of the cell M_(m-1)3, connected in parallel with the second nMOS transistor Q_(m-1)32 , wherein an output node connecting the source electrode of the first nMOS transistor Q_(m-1)31 and the drain electrode of the second nMOS transistor Q_(m-1)32 serves as an output terminal of the cell M_(m-1)3, configured to deliver the signal stored in the capacitor C_(m-1)3 to the next bit-level cell. The third-column cell M_m3 on the m-th row encompasses a first nMOS transistor Q_m31 having a drain electrode connected to the clock signal supply line through a first delay element D_m31 and a gate electrode connected to the output terminal of the previous bit-level cell M_m2 through a second delay element D_m32; a second nMOS transistor Q_m32 having a drain electrode connected to a source electrode of the first nMOS transistor Q_m31, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_m3configured to store the information of the cell M_m3, connected in parallel with the second nMOS transistor Q_m32 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_m31 and the drain electrode of the second nMOS transistor Q_m32 serves as an output terminal of the cell M_m3, configured to deliver the signal stored in the capacitor C_m3 to the next bit-level cell.

Still furthermore, as illustrated in Fig.4, the n-th memory unitof word-size level is implemented by a vertical array of bit-level cell M_1n, M_2n, M_3n, ........, M_m-1,n, M_mn in the n-th column of the m * n matrix. The n-th-column cell M_1n on the first row encompasses a first nMOS transistor Q_1n1 having a drain electrode connected to the clock signal supply line through a first delay element D_1n1 and a gate electrode connected to the bit-level output terminal of the previous bit-level cell M_1(n-1) through a second delay element D_1n2; a second nMOS transistor Q_1n2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_1n1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_1nconfigured to store the information of the cell M_1n, connected in parallel with the second nMOS transistor Q_1n2 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_1n1 and the drain electrode of the second nMOS transistor Q_1n2 serves as a bit-level output terminal of the cell M_1n, configured to deliver the signal stored in the capacitor C_1n to a first bit-level output terminal. The n-th-column cell M_2n on the second row encompasses a first nMOS transistor Q_2n1 having a drain electrode connected to the clock signal supply line through a first delay element D_2n1 and a gate electrode connected to the bit-level output terminal of the previous bit-level cell M_2(n-1) through a second delay element D_2n2; a second nMOS transistor Q_2n2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_2n1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_2nconfigured to store the information of the cell M_2n, connected in parallel with the second nMOS transistor Q_2n2 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_2n1 and the drain electrode of the second nMOS transistor Q_2n2 serves as a bit-level output terminal of the cell M_2n, configured to deliver the signal stored in the capacitor C_2n to a second bit-level output terminal. The n-th-column cell M_3n on the third row encompasses a first nMOS transistor Q_3n1 having a drain electrode connected to the clock signal supply line through a first delay element D_3n1 and a gate electrode connected to the bit-level output terminal of the previous bit-level cell M_3(n-1) through a second delay element D_3n2; a second nMOS transistor Q_3n2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_3n1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_3nconfigured to store the information of the cell M_3n, connected in parallel with the second nMOS transistor Q_3n2 , wherein an output node connecting the source electrode of the first nMOS transistor Q_3n1 and the drain electrode of the second nMOS transistor Q_3n2 serves as a bit-level output terminal of the cell M_3n, configured to deliver the signal stored in the capacitor C_3n to a third bit-level output terminal. ............ The n-th-column cell M_(m-1)n on the (m-1)-th row encompasses a first nMOS transistor Q_(m-1)n1 having a drain electrode connected to the clock signal supply line through a first delay element D_(m-1)n1 and a gate electrode connected to the bit-level output terminal of the previous bit-level cell M_{(m-1) (n-1)} through a second delay element D_(m-1)n2; a second nMOS transistor Q_(m-1)n2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(m-1)n1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(m-1)nconfigured to store the information of the cell M_(m-1)n, connected in parallel with the second nMOS transistor Q_(m-1)n2 , wherein an output node connecting the source electrode of the first nMOS transistor Q_(m-1)n1 and the drain electrode of the second nMOS transistor Q_(m-1)n2 serves as a bit-level output terminal of the cell M_(m-1)n, configured to deliver the signal stored in the capacitor C_(m-1)n to a (m-1)-th bit-level output terminal. The n-th-column cell M_mn on the m-th row encompasses a first nMOS transistor Q_mn1 having a drain electrode connected to the clock signal supply line through a first delay element D_mn1 and a gate electrode connected to the bit-level output terminal of the previous bit-level cell M_m(n-1) through a second delay element D_mn2; a second nMOS transistor Q_mn2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_mn1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_mnconfigured to store the information of the cell M_mn, connected in parallel with the second nMOS transistor Q_mn2 ,wherein an output node connecting the source electrode of the first nMOS transistor Q_mn1 and the drain electrode of the second nMOS transistor Q_mn2 serves as a bit-level output terminal of the cell M_mn, configured to deliver the signal stored in the capacitor C_mn to a m-th bit-level output terminal.

As illustrated in Fig. 5, a bit-level cell M_ij of the j-th column and on the i-th row, in the representative 2 * 2 cell-array of the marching main memory used in the computer system pertaining to the first embodiment of the present invention, encompasses a first nMOS transistor Q_ij1 having a drain electrode connected to a clock signal supply line through a first delay element D_ij1 and a gate electrode connected to the output terminal of the previous bit-level cell through a second delay element D_ij2; a second nMOS transistor Q_ij2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_ij1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_ijconfigured to store the information of the bit-level cell M_ij, connected in parallel with the second nMOS transistor Q_ij2 , wherein an output node connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2 serves as an output terminal of the bit-level cell M_ij, configured to deliver the signal stored in the capacitor C_ij to the next bit-level cell M_i(j+1).

A column bit-level cell M_i(j+1) of the (j+1)-th column and on the i-th row encompasses a first nMOS transistor Q_i(j+1)1 having a drain electrode connected to clock signal supply line through a first delay element D_i(j+1)1 and a gate electrode connected to the output terminal of the previous bit-level cell M_ij through a second delay element D_i(j+1)2; a second nMOS transistor Q_i(j+1)2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_i(j+1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i(j+1)configured to store the information of the bit-level cell M_i(j+1), connected in parallel with the second nMOS transistor Q_i(j+1)2, wherein an output node connecting the source electrode of the first nMOS transistor Q_i(j+1)1 and the drain electrode of the second nMOS transistor Q_i(j+1)2 serves as an output terminal of the bit-level cell M_i(j+1), configured to deliver the signal stored in the capacitor C_i(j+1) to the next cell.

And, a bit-level cell M_(i+1)j of the j-th column and on the (i+1)-th row encompasses a first nMOS transistor Q_(i+1)j1 having a drain electrode connected to the clock signal supply line through a first delay element D_(i+1)j1 and a gate electrode connected to the output terminal of the previous bit-level cell through a second delay element D_(i+1)j2; a second nMOS transistor Q_(i+1)j2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(i+1)j1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(i+1)jconfigured to store the information of the bit-level cell M_(i+1)j, connected in parallel with the second nMOS transistor Q_(i+1)j2 , wherein an output node connecting the source electrode of the first nMOS transistor Q_(i+1)j1 and the drain electrode of the second nMOS transistor Q_(i+1)j2 serves as an output terminal of the bit-level cell M_(i+1)j, configured to deliver the signal stored in the capacitor C_(i+1)j to the next bit-level cell M_(i+1)(j+1).

Furthermore, a bit-level cell M_(i+1)(j+1) of the (j+1)-th column and on the (i+1)-th row encompasses a first nMOS transistor Q_(i+1)(j+1)1 having a drain electrode connected to the clock signal supply line through a first delay element D_(i+1)(j+1)1 and a gate electrode connected to the output terminal of the previous bit-level cell M_(i+1)j through a second delay element D_(i+1)(j+1)2; a second nMOS transistor Q_(i+1)(j+1)2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_(i+1)(j+1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_(i+1)(j+1)configured to store the information of the bit-level cell M_(i+1)(j+1), connected in parallel with the second nMOS transistor Q_(i+1)(j+1)2, wherein an output node connecting the source electrode of the first nMOS transistor Q_(i+1)(j+1)1 and the drain electrode of the second nMOS transistor Q_(i+1)(j+1)2 serves as an output terminal of the bit-level cell M_(i+1)(j+1), configured to deliver the signal stored in the capacitor C_(i+1)(j+1) to the next cell.

As illustrated in Fig. 6, the j-th bit-level cell M_ij on the i-th row encompasses a first nMOS transistor Q_ij1 having a drain electrode connected to a clock signal supply line through a first delay element D_ij1 and a gate electrode connected to the output terminal of the previous cell through a second delay element D_ij2; a second nMOS transistor Q_ij2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_ij1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_ijconfigured to store the information of the bit-level cell M_ij, connected in parallel with the second nMOS transistor Q_ij2 .

In the circuit configuration illustrated in Fig. 6, the second nMOS transistor Q_ij2 serves as a reset-transistor configured to reset the signal charge stored in the capacitor C_ij, when a clock signal of high-level (or a logical level of "1") is applied to the gate electrode of the second nMOS transistor Q_ij2, discharging the signal charge already stored in the capacitor C_ij.

Figs. 7A and 7B illustrate a schematic example of the transistor-level responses of the bit-level cell M_ij illustrated in Fig. 6, which is one of the bit-level cells used in the computer system pertaining to the first embodiment of the present invention, to a waveform of a clock signal illustrated by broken line. The clock signal illustrated by broken line swings periodically between the logical levels of "1" and "0" with the clock period TAU(Greek-letter) _clock. In Figs. 7A and 7B, t₁-t₀ (= t₂-t₁ = t₃-t₂ = t₄-t₃) is defined to be a quarter of the clock period TAU_clock(=TAU_clock /4).

(a) As illustrated in Fig. 7A(a), at time "t₀", although the clock signal of high-level illustrated by the broken line is applied both to a drain electrode of the first nMOS transistor Q_ij1 through a first ideal delay element D_ij1 and to a gate electrode of the second nMOS transistor Q_ij2, the second nMOS transistor Q_ij2 keeps off-state until the first nMOS transistor Q_ij1will establish on-state at time "t₁", because the potential of the output node N_out, connecting between a source electrode of the first nMOS transistor Q_ij1 and a drain electrode of the second nMOS transistor Q_ij2, is supposed to be a floating state, lying between the logical levels of "0" and "1", between the time "t₀" and the time "t₁".

(b) Owing to the first ideal delay element D_ij1, because the turn on of the first nMOS transistor Q_ij1is delayed by t₁-t₀ =TAU_clock /4, the first nMOS transistor Q_ij1 becomes active as a transfer-transistor at time "t₁", and the potential of the output node N_out becomes the logical level "1". Here, it is assumed that the first ideal delay element D_ij1 can achieve a delay of TAU_clock /4 with very sharp leading edge, by which the rise time can be neglected. That is, as illustrated by solid line with very sharp leading edge and very sharp trailing edge in Fig. 7A(a), the clock signal applied at time "t₀" is delayed by t₁-t₀ =TAU_clock /4. Then, as illustrated in Fig. 7A(c)-(d), if the signal stored in the previous bit-level cell M_i(j-1) is the logical level of "1", the second nMOS transistor Q_ij2 becomes active as a reset-transistor, and any signal charge stored in the capacitor C_ij is driven to be discharged, and at time "t₂".

(c) The first nMOS transistor Q_ij1 becomes completely active as the transfer-transistor at time "t₂", delayed by a predetermined delay time t_d2 = t₂-t₀ =TAU_clock /2, determined by the second ideal delay element D_ij2. Here, it is assumed that the second ideal delay element D_ij2 can achieve a delay of TAU_clock /2 with very sharp leading edge, by which the rise time can be neglected. Then, if the signal of the logical level of "1" stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) on the i-th row to the gate electrode of the first nMOS transistor Q_ij1, at time "t₂", the signal charge stored in the capacitor C_ij is completely discharged to establish the logical level of "0", as illustrated in Fig. 7A(b), and the first nMOS transistor Q_ij1 begins transferring the signal of the logical level of "1" stored in the previous bit-level cell M_i(j-1), to the capacitor C_ij so as to execute marching AND-gate operation as illustrated in Fig. 7A(c)-(d). That is, with an input signal of "1" provided by the clock signal and another input signal of "1" provided by the previous bit-level cell M_i(j-1), the conventional 2-input AND operation of:

1 + 1= 1

can be executed. By the way, if the signal charge stored in the capacitor C_ij is of the logical level of "1", the capacitor C_ij can begin discharging at time "t₀", because the second nMOS transistor Q_ij2 can become active as the reset-transistor with the clock signal of the high-level illustrated by the broken line applied to the gate electrode of the second nMOS transistor Q_ij2 at time "t₀", if the operation of the second nMOS transistor Q_ij2 has no delay.

(d) Alternatively, as illustrated in Fig. 7B(c)-(d), if the signal stored in the previous bit-level cell M_i(j-1) is the logical level of "0", the first nMOS transistor Q_ij1 keeps off-sate at any time "t₀", "t₁", "t₂" and "t₃". As above-mentioned, if the signal charge stored in the capacitor C_ij is of the logical level of "1", although the first nMOS transistor Q_ij1 keeps off-sate, the capacitor C_ij can begin discharging at time "t₀", because the second nMOS transistor Q_ij2 can become active as the reset-transistor with the clock signal of the high-level illustrated by the broken line applied to the gate electrode of the second nMOS transistor Q_ij2 at time "t₀", and the marching AND-gate operation of:

1 + 0 = 0

is executed as illustrated in Fig. 7A(c)-(d), with an input signal of "1" provided by the clock signal and another input signal of "0" provided by the previous bit-level cell M_i(j-1). However, if the signal charge stored in the capacitor C_ij is of of the logical level of "0", because both of the first nMOS transistor Q_ij1 the second nMOS transistor Q_ij2 keep the off-sate, the capacitor C_ij keep the logical level of "0" any time "t₀", "t₁", "t₂"and "t₃", and the marching AND-gate operation of is executed as illustrated in Fig. 7A(c)-(d). The output node N_out connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2 serves as an output terminal of the bit-level cell M_ij, and the output terminal of the bit-level cell M_ij delivers the signal stored in the capacitor C_ij to the next bit-level cell on the i-th row.

Furthermore, Fig. 7C illustrates an actual example of the response to the waveform of the clock signal, for a case that both of the first delay element D_ij1 and the second delay element D_ij2 are implemented by R-C delay circuit, as illustrated in Fig. 8. In a normal operation of the marching memory, the signal charge stored in the capacitor C_ij is actually either of the logical level of "0" or"1", and if the signal charge stored in the capacitor C_ij is of the logical level of "1", although the first nMOS transistor Q_ij1 still keeps off-sate, the capacitor C_ij can begin discharging at time "t₀", because the second nMOS transistor Q_ij2 can become active when the clock signal of the high-level is applied to the gate electrode of the second nMOS transistor Q_ij2, if an ideal operation of the second nMOS transistor Q_ij2 with no delay can be approximated. Therefore, if the signal charge stored in the capacitor C_ij is actually of the logical level of "1", after the clock signal of high-level has been applied to the gate electrode of the second nMOS transistor Q_ij2 and the signal charge stored in the capacitor C_ij has been discharged, the first nMOS transistor Q_ij1 becomes active as a transfer-transistor, delayed by a predetermined delay time t_d1 determined by the first delay element D_ij1 implemented by the R-C delay circuit. And when the signal stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) on the i-th row to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 transfers the signal stored in the previous bit-level cell M_i(j-1), further delayed by a predetermined delay time t_d2 determined by the second delay element D_ij2 to the capacitor C_ij. An output node N_out connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2 serves as an output terminal of the bit-level cell M_ij, and the output terminal of the bit-level cell M_ij delivers the signal stored in the capacitor C_ij to the next bit-level cell on the i-th row.

As illustrated in Fig. 7C, the clock signal swings periodically between the logical levels of "1" and "0", with a predetermined clock period (clock cycle time) TAU_clock, and when the clock signal becomes the logical level of "1", the second nMOS transistor Q_ij2 begins to discharge the signal charge, which is already stored in the capacitor C_ij at a previous clock cycle. And, after the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_ij is completely discharged to the potential of the logical level of "0", the first nMOS transistor Q_ij1 becomes active as the transfer-transistor, delayed by the predetermined delay time t_d1 determined by the first delay element D_ij1. The delay time t_d1 may be set to be equal to 1/4TAU_clock preferably. Thereafter, when the signal stored in the previous bit-level cell M_i(j-1) on the i-th row is fed from the previous bit-level cell M_i(j-1) to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 transfers the signal stored in the previous bit-level cell M_i(j-1), further delayed by the predetermined delay time t_d2 determined by the second delay element D_ij2 implemented by the R-C delay circuit to the capacitor C_ij.

For example, if the logical level of "1" stored in the previous bit-level cell M_i(j-1) on the i-th row is fed from the previous bit-level cell M_i(j-1) to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 becomes conductive state, and the logical level of "1" is stored in the capacitor C_ij. On the other hand, if the logical level of "0" stored in the previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 keeps cut-off state, and the logical level of "0" is maintained in the capacitor C_ij. Therefore, the bit-level cell M_ij can establish "a marching AND-gate" operation. The delay time t_d2 shall be longer than the delay time t_d1, and the delay time t_d2 may be set to be equal to 1/2TAU_clock preferably.

Because the clock signal swings periodically between the logical levels of "1" and "0", with the clock period TAU_clock, then, the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, and the output node N_out connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2 cannot deliver the signal transferred from the previous bit-level cell M_i(j-1) further to the next bit-level cell M_i(j+1) at a time when time proceeds 1/2TAU_clock, as the signal is blocked to be transferred to the gate electrode of the next first nMOS transistor Q_i(j+1)1 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D_i(j+1)2. When the clock signal becomes the logical level of "1" again at a time when time proceeds TAU_clock, the output node N_out connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2, which is serving as the output terminal of the bit-level cell M_ij, can deliver the signal stored in the capacitor C_ij to the next bit-level cell M_i(j+1) at the next clock cycle.

Turning back to Fig.4, when the clock signal illustrated in Fig. 7A (a) or Fig. 7C becomes the logical level of "1", a sequence of the second nMOS transistors Q₁₁₂, Q₂₁₂, Q₃₁₂, ........, Q_m-1,12, Q_m12 in the first memory unit U₁ begin to discharge the signal charges, respectively, which are already stored in the capacitors C₁₁, C₂₁, C₃₁, ........, C_m-1,1, C_m1, respectively, in the first memory unit U₁ at a previous clock cycle. And, after the clock signal of the logical level of "1" is applied to the gate electrodes of the sequence of the second nMOS transistors Q₁₁₂, Q₂₁₂, Q₃₁₂, ........, Q_m-1,12, Q_m12, respectively, and the signal charges stored in the capacitors C₁₁, C₂₁, C₃₁, ........, C_m-1,1, C_m1 are completely discharged to the potential of the logical level of "0", a sequence of the first nMOS transistors Q₁₁₁, Q₂₁₁, Q₃₁₁, ........, Q_m-1,11, Q_m11 becomes active as the transfer-transistors, delayed by the delay time t_d1 determined by the first delay elements D₁₁₁, D₂₁₁, D₃₁₁, ........, D_m-1,11, D_m11, respectively. Thereafter, when a sequence of signals of word size, whish is multiples of eight bits, such as 16, 32, and 64 bits are entered to the gate electrodes of the sequence of the first nMOS transistors Q₁₁₁, Q₂₁₁, Q₃₁₁, ........, Q_m-1,11, Q_m11, the sequence of the first nMOS transistors Q₁₁₁, Q₂₁₁, Q₃₁₁, ........, Q_m-1,11, Q_m11 transfer the sequence of signals of word size to the capacitors C₁₁, C₂₁, C₃₁, ........, C_m-1,1, C_m1,delayed by the delay time t_d2 determined by the second delay elements D₁₁₂, D₂₁₂, D₃₁₂, ........, D_m-1,12, D_m12, respectively.

When the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, each of the output nodes connecting the source electrodes of the first nMOS transistors Q₁₁₁, Q₂₁₁, Q₃₁₁, ........, Q_m-1,11, Q_m11 and the drain electrodes of the second nMOS transistors Q₁₁₂, Q₂₁₂, Q₃₁₂, ........, Q_m-1,12, Q_m12 cannot deliver the signals, which are entered to the gate electrodes of the first nMOS transistors Q₁₁₁, Q₂₁₁, Q₃₁₁, ........, Q_m-1,11, Q_m11, further to the next bit-level cell M₁₂, M₂₂, M₃₂, ........, M_m-1,2, M_m2 at a time when time proceeds 1/2TAU_clock, as each of the signals is blocked to be transferred to the gate electrodes of the next first nMOS transistors Q₁₂₁, Q₂₂₁, Q₃₂₁, ........, Q_m-1,21, Q_m21 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D₁₂₂, D₂₂₂, D₃₂₂, ........, D_m-1,22, D_m22.

And, at a time when time proceeds TAU_clock, when the next clock signal becomes the logical level of "1" again, a sequence of the second nMOS transistors Q₁₂₂, Q₂₂₂, Q₃₂₂, ........, Q_m-1,22, Q_m22 in the second memory unit U₂ begin to discharge the signal charges, respectively, which are already stored in the capacitors C₁₂, C₂₂, C₃₂, ........, C_m-1,2, C_m2, respectively, in the second memory unit U₂ at the previous clock cycle. And, after the clock signal of the logical level of "1" is applied to the gate electrodes of the sequence of the second nMOS transistors Q₁₂₂, Q₂₂₂, Q₃₂₂, ........, Q_m-1,22, Q_m22, respectively, and the signal charges stored in the capacitors C₁₂, C₂₂, C₃₂, ........, C_m-1,2, C_m2, are completely discharged to the potential of the logical level of "0", a sequence of the first nMOS transistors Q₁₂₁, Q₂₂₁, Q₃₂₁, ........, Q_m-1,21, Q_m21 becomes active as the transfer-transistors, delayed by the delay time t_d1 determined by the first delay elements D₁₂₁, D₂₂₁, D₃₂₁, ........, D_m-1,21, D_m21, respectively. Thereafter, when the sequence of signals of word size stored in the previous capacitors C₁₁, C₂₁, C₃₁, ........, C_m-1,1, C_m1 are fed to the gate electrode of the sequence of the first nMOS transistors Q₁₂₁, Q₂₂₁, Q₃₂₁, ........, Q_m-1,21, Q_m21, and the first nMOS transistor Q₁₂₁, Q₂₂₁, Q₃₂₁, ........, Q_m-1,21, Q_m21 transfer the sequence of signals of word size, delayed by the delay time t_d2 determined by the second delay element D₁₂₂, D₂₂₂, D₃₂₂, ........, D_m-1,22, D_m22, to the capacitors C₁₂, C₂₂, C₃₂, ........, C_m-1,2, C_m2.

When the clock signal becomes the logical level of "0" at a time when time further proceeds (1+1/2)TAU_clock, each of the output nodes connecting the source electrodes of the first nMOS transistors Q₁₂₁, Q₂₂₁, Q₃₂₁, ........, Q_m-1,21, Q_m21 and the drain electrodes of the second nMOS transistors Q₁₂₂, Q₂₂₂, Q₃₂₂, ........, Q_m-1,22, Q_m2 cannot deliver the signals stored in the previous bit-level cell M₁₁, M₂₁, M₃₁, ........, M_m-1,1, M_m1 further to the next bit-level cell M₁₂, M₂₂, M₃₂, ........, M_m-1,2, M_m2 at a time when time proceeds (1+1/2)TAU_clock, as each of the signals is blocked to be transferred to the gate electrode of the next first nMOS transistor Q₁₃₁, Q₂₃₁, Q₃₃₁, ........, Q_m-1,31, Q_m31 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D₁₃₂, D₂₃₂, D₃₃₂, ........, D_m-1,32, D_m32.

And, at a time when time further proceeds 2TAU_clock, when the next clock signal becomes the logical level of "1" again, a sequence of the second nMOS transistors Q₁₃₂, Q₂₃₂, Q₃₃₂, ........, Q_m-1,32, Q_m32 in the third memory unit U₃ begin to discharge the signal charges, respectively, which are already stored in the capacitors C₁₃, C₂₃, C₃₃, ........, C_m-1,3, C_m3, respectively, in the third memory unit U₃ at the previous clock cycle. And, after the clock signal of the logical level of "1" is applied to the gate electrodes of the sequence of the second nMOS transistors Q₁₃₂, Q₂₃₂, Q₃₃₂, ........, Q_m-1,32, Q_m32, respectively,and the signal charges stored in the capacitors C₁₃, C₂₃, C₃₃, ........, C_m-1,3, C_m3, are completely discharged to the potential of the logical level of "0", a sequence of the first nMOS transistors Q₁₃₁, Q₂₃₁, Q₃₃₁, ........, Q_m-1,31, Q_m31 becomes active as the transfer-transistors, delayed by the delay time t_d1 determined by the first delay elements D₁₃₁, D₂₃₁, D₃₃₁, ........, D_m-1,31, D_m31, respectively. Thereafter, when the sequence of signals of word size stored in the previous capacitors C₁₂, C₂₂, C₃₂, ........, C_m-1,2, C_m2 are fed to the gate electrode of the sequence of the first nMOS transistors Q₁₃₁, Q₂₃₁, Q₃₃₁, ........, Q_m-1,31, Q_m31, and the first nMOS transistor Q₁₃₁, Q₂₃₁, Q₃₃₁, ........, Q_m-1,31, Q_m31 transfer the sequence of signals of word size, delayed by the delay time t_d2 determined by the second delay element D₁₃₂, D₂₃₂, D₃₃₂, ........, D_m-1,32, D_m32, to the capacitors C₁₃, C₂₃, C₃₃, ........, C_m-1,3, C_m3.

As illustrated in Fig. 8, each of the first delay element D_ij1 and the second delay element D_ij2 can be implemented by known "resistive-capacitive delay" or "R-C delay". In the RC circuit, the value of the time constant (in seconds) is equal to the product of the circuit resistance (in ohms) and the circuit capacitance (in farads), i.e. t_d1 , t_d2 = R * C. Because the structure of the RC circuit is very simple, it is preferable to use the RC circuit for the first delay element D_ij1 and the second delay element D_ij2. However, the RC circuit is mere example, and the first delay element D_ij1 and the second delay element D_ij2.can be implemented by another passive delay elements, or various active delay element, which may include active element of transistor, etc.

Fig.9 illustrates an example of the top view of the actual planar pattern of the bit-level cell M_ij of the j-th column and on the i-th row illustrated in Fig. 8, which has the first delay element D_ij1 and the second delay element D_ij2 implemented by the R-C delay circuit, and Fig.10 illustrates the corresponding cross-sectional view taken on the line A-A of Fig.9. As illustrated in Fig.9, the first delay element D_ij1 is implemented by a first meandering line 91 of conductive wire, and the second delay element D_ij2 is implemented by a second meandering line 97 of conductive wire.

In Fig. 9, the first nMOS transistor Q_ij1 has a drain electrode region 93 connected to the first meandering line 91 via a contact plug 96a. The other end of the first meandering line 91 opposite to the end connected to the drain electrode region 93 of the first nMOS transistor Q_ij1 is connected to the clock signal supply line. The drain electrode region 93 is implemented by an n⁺ semiconductor region. A gate electrode of the first nMOS transistor Q_ij1 is implemented by the second meandering line 97. The other end of the second meandering line 97 opposite to the end serving as the gate electrode of the first nMOS transistor Q_ij1 is connected to the output terminal of the previous cell.

The second nMOS transistor Q_ij2 has a drain electrode region implemented by a common n⁺ semiconductor region 94, which also serves as the source electrode region of the first nMOS transistor Q_ij1, a gate electrode 98 connected to the clock signal supply line via a contact plug 96a, and a source electrode region 95 connected to the ground potential via a contact plug 96a. The source electrode region 95 is implemented by an n⁺ semiconductor region. Because the common n⁺ semiconductor region 94 is the output node connecting the source electrode region of the first nMOS transistor Q_ij1 and the drain electrode region of the second nMOS transistor Q_ij2, the common n⁺ semiconductor region 94 is connected to a surface wiring 92b via a contact plug 96d. The common n⁺ semiconductor region 94 serves as the output terminal of the bit-level cell M_ij, and delivers the signal stored in the capacitor C_ij to the next bit-level cell through the surface wiring 92b.

As illustrated in Fig. 10, the drain electrode region 93, the common n⁺ semiconductor region 94, and the source electrode region 95 is provided at the surface of and in the upper portion of the p-type semiconductor substrate 81. Instead of the p-type semiconductor substrate 81, the drain electrode region 93, the common n⁺ semiconductor region 94, and the source electrode region 95 can be provided in the upper portion of the p-well, or p-type epitaxial layer grown on a semiconductor substrate. On the p-type semiconductor substrate 81, an element isolation insulator 82 is provided so as to define an active area of the p-type semiconductor substrate 81 as a window provided in the element isolation insulator 82. And the drain electrode region 93, the common n⁺ semiconductor region 94, and the source electrode region 95 is provided in the active area, surrounded by the element isolation insulator 82. At the surface of and on the active area, a gate insulating film 83 is provided. And the gate electrode of the first nMOS transistor Q_ij1 implemented by the second meandering line 97 and the gate electrode 98 of the second nMOS transistor Q_ij2 are provided on the gate insulating film 83.

As illustrated in Fig. 10, a first interlayer dielectric film 84 is provided on the second meandering line 97 and the gate electrode 98. On a part of the first interlayer dielectric film 84, a bottom electrode 85 of the capacitor C_ijconfigured to store the information of the bit-level cell M_ij is provided. The bottom electrode 85 is made of conducting film, and a contact plug 96c is provided in the first interlayer dielectric film 84 so as to connect between the bottom electrode 85 and the source electrode region 95. And, on the bottom electrode 85, a capacitor insulating film 86 is provided.

Furthermore, on the capacitor insulating film 86, a top electrode 87 of the capacitor C_ijis provided so as to occupy an upper portion of the bottom electrode 85. The top electrode 87 is made of conducting film. Although the illustration is omitted in the cross-sectional view illustrated in Fig. 10, the top electrode 87 is electrically connected to the common n⁺ semiconductor region 94 so as to establish an electric circuit topology that the capacitor C_ijis connected in parallel with the second nMOS transistor Q_ij2. A variety of insulator films may be used as the capacitor insulating film 86. The miniaturized marching main memory may be required to occupy a small area of the bottom electrode 85 opposing the top electrode 87. However, to allow the marching main memory to function successfully, the capacitance between the bottom electrode 85 and the top electrode 87 via the capacitor insulating film 86 needs to maintain a constant value. In particular, with a miniaturized marching main memory with a minimum line width of approximately 100 nm or less, usage of a material with a dielectric constant e_r greater than that of a silicon oxide (SiO₂) film is preferred, considering the storage capacitance between the bottom electrode 85 and the top electrode 87. With an ONO film, for example, the ratio in thickness of the upper layer silicon oxide film, the middle layer silicon nitride film, and the underlayer silicon oxide film is selectable, however, a dielectric constant e_r of approximately 5 to 5.5 can be provided. Alternatively, a single layer film made from any one of a strontium oxide (SrO) film with e_r = 6, a silicon nitride (Si₃N₄) film with e_r = 7, an aluminum oxide (Al₂O₃) film where e_r = 8 - 11, a magnesium oxide (MgO) film where e_r = 10, an yttrium oxide (Y₂O₃) film where e_r = 16-17, a hafnium oxide (HfO₂) film where e_r = 22-23, a zirconium oxide (ZrO₂) film where e_r = 22-23, a tantalum oxide (Ta₂O₅) film where e_r = 25- 27, or a bismuth oxide (Bi₂O₃) film where e_r = 40, or a composite film embracing at least two of these plural layers thereof may be used. Ta₂O₅ and Bi₂O₃ show disadvantages in lacking thermal stability at the interface with the polysilicon. Furthermore, it may be a composite film made from a silicon oxide film and these films. The composite film may have a stacked structure of triple-levels or more. In other words, it should be an insulating film containing a material with the relative dielectric constant e_r of 5 to 6 or greater in at least a portion thereof. However, in the case of a composite film, selecting a combination that results in having an effective relative dielectric constant e_reff of 5 to 6 or greater measured for the entire film is preferred. Moreover, it may also be an insulating film made from an oxide film of a ternary compound such as a hafnium aluminate (HfAlO) film.

Furthermore, a second interlayer dielectric film 87 is provided on the top electrode 87. And the first meandering line 91 is provided on second interlayer dielectric film 87. As illustrated in Fig. 10, the contact plug 96a is provided, penetrating the first interlayer dielectric film 84, the capacitor insulating film 86 and the second interlayer dielectric film 87 so as to connect between the first meandering line 91 and the drain electrode region 93

In a topology illustrated in Figs.9 and 10, the capacitance C of the R-C delay is implemented by the stray capacitance associated with the first meandering line 91 and the second meandering line 97. Because both R and C are proportional to wire lengths of the first meandering line 91 and the second meandering line 97, the delay times t_d1, t_d2 can be easily designed by electing the wire lengths of the first meandering line 91 and the second meandering line 97. Furthermore, we can design the thickness, the cross section, or the resistivity of the first meandering line 91 and the second meandering line 97 to as to achieve desired value of the delay times t_d1, t_d2.

For example, because the delay time t_d2 shall be twice of the delay time t_d1, the wire length of the second meandering line 97 can be designed as 2^1/2 time of the wire length the first meandering line 91, if we use the same thickness, the same cross section, and the material having the same specific resistively for the first meandering line 91 and the second meandering line 97, and further the same effective thickness and the same effective dielectric constant for the insulating film implementing the stray capacitance for the R-C delay (= R * C). However, if we use different materials for the first meandering line 91 and the second meandering line 97, the wire lengths of the first meandering line 91 and the second meandering line 97shall be determined depending on the resistivities of the first meandering line 91 and the second meandering line 97 so as to achieve the required values of the delay times t_d1, t_d2. For example, in a case that the second meandering line 97 is formed of polycrystalline silicon, and the first meandering line 91 is formed of refractory material such as tungsten (W), molybdenum (Mo), platinum (Pt), having a higher resistivity than the polycrystalline silicon, the wire lengths of the first meandering line 91 and the second meandering line 97 are determined depending on the resistivities of the first meandering line 91 and the second meandering line 97 so as to achieve the required values of the delay times t_d1, t_d2.

Furthermore, although the first meandering line 91 and the second meandering line 97 are illustrated in Fig.9, the illustrated meandering topology for resistor R is mere example, and other topologies such as a straight line configuration can be used depending upon the required values of resistor R and capacitance C. In a very high speed operation of the marching main memory 31, the delineation of extrinsic resistor elements R can be omitted, if parasitic resistance (stray resistance) and parasitic capacitance (stray capacitance) can achieve the required delay times t_d1, t_d2.

In the configuration illustrated in Figs.4-6, although an isolation between a signal-storage state of the (j-1)-th bit-level cell M_ij-1 on the i-th row and a signal-storage state of the j-th bit-level cell M_ij on the i-th row can be established by a propagation delay accompanying the signal propagation path between the output terminal of the (j-1)-th bit-level cell M_ij-1 and the gate electrode of the first nMOS transistor Q_ij1 of the j-th bit-level cell M_ij, the propagation delay is mainly ascribable to the value of the second delay element D_ij2, it is preferable to insert an inter-unit cell B_ij between the (j-1)-th bit-level cell M_ij-1 and the j-th bit-level cell M_ij, as illustrated in Figs. 11 and 13.

Although the inter-unit cell B_ij is provided so as to isolate the signal-storage state of the j-th bit-level cell M_ij in the j-th memory unit U_j from the signal-storage state of the (j-1)-th bit-level cell M_ij-1 in the (j-1)-th memory unit U_j-1, the inter-unit cell B_ij transfers a signal from the (j-1)-th bit-level cell M_ij-1 to the j-th bit-level cell M_ij at a required timing determined by a clock signal, which is supplied through the clock signal supply line. Because the j-th memory unit U_j stores information of byte size or word size by the sequence of bit-level cells arrayed in the j-th memory unit U_j, and the (j-1)-th memory unit U_j-1 stores information of byte size or word size by the sequence of bit-level cells arrayed in the (j-1)-th memory unit U_j-1, a sequence of inter-unit cells arrayed in parallel with the memory units U_j-1 and U_j transfers the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line so that the information of byte size or word size can march along a predetermined direction, pari passu. As illustrated in Figs. 11 and 13, because the input terminal of the j-th bit-level cell M_ij on the i-th row is connected to the inter-unit cell B_ij, the signal charge stored in the (j-1)-th bit-level cell M_ij-1 is fed to the second delay element D_ij2 through the inter-unit cell B_ij at the required timing, and the transfer operation of the signal charge is cut off at periods other than the required timing.

In Figs. 11 and 13, although an example of the inter-unit cell B_ij, which encompasses a single isolation transistor Q_ij3 having a first main-electrode connected to the output terminal of the (j-1)-th bit-level cell M_ij, a second main-electrode connected to the input terminal of the j-th bit-level cell M_ij and a control electrode connected to the clock signal supply line, the structure of the inter-unit cell B_ij is not limited to the configuration illustrated in Figs. 11 and 13. For example, the inter-unit cell B_ij may be implemented by a clocked-circuit having a plurality of transistors, which can transfer the signal from the (j-1)-th bit-level cell M_ij-1 to the j-th bit-level cell M_ij at the required timing determined by the clock signal.

Similar to the configuration illustrated in Fig. 5, the j-th bit-level cell M_ij encompasses the first nMOS transistor Q_ij1 having the drain electrode connected to the clock signal supply line through the first delay element D_ij1 and the gate electrode connected to the inter-unit cell B_ij through the second delay element D_ij2; the second nMOS transistor Q_ij2 having the drain electrode connected to the source electrode of the first nMOS transistor Q_ij1, the gate electrode connected to the clock signal supply line, and the source electrode connected to the ground potential; and the capacitor C_ijconfigured to store the information of the bit-level cell M_ij, connected in parallel with the second nMOS transistor Q_ij2.

An example of planar structure of the inter-unit cell B_ij, encompassing a single isolation transistor Q_ij3 of nMOS transistor is illustrated in Fig. 12, in addition to the configuration of the bit-level cell M_ij, which are alreadyillustrated in Fig. 9. In the bit-level cell M_ij, the first nMOS transistor Q_ij1 having the drain electrode region 93, the first meandering line 91 connected to the drain electrode region 93 via a contact plug 96a, the second meandering line 97 implementing the gate electrode of the first nMOS transistor Q_ij1, and the second nMOS transistor Q_ij3 having the drain electrode region implemented by the common n⁺ semiconductor region 94, serving as the output terminal of the bit-level cell M_ij are illustrated.

In Fig. 12, the isolation transistor Q_ij3 of the inter-unit cell B_ij has a first main-electrode region implemented by a left side of an n⁺ semiconductor region 90, a gate electrode 99 connected to the clock signal supply line, and a second main-electrode region implemented by a right side of the n⁺ semiconductor region 90. The second main-electrode region is connected to one end of the second meandering line 97 opposite to the other end of the second meandering line 97, which serves as the gate electrode of the first nMOS transistor Q_ij1 via a contact plug 96e, and first main-electrode region is connected to the output terminal of the previous cell M_ij-1 via a contact plug 96f. Although the illustration is omitted, similar to the structure illustrated in Fig. 10, on an interlayer dielectric film provided on the second meandering line 97, for example, a parallel plate structure of the capacitor C_ijconfigured to store the information of the bit-level cell M_ij may be provided, being connected in parallel with the second nMOS transistor Q_ij2.

In Fig. 13, in addition to the configuration illustrated in Fig. 11, another inter-unit cell B_i(j-1) is provided between the (j-2)-th bit-level cell M_i(j-2) and the (j-1)-th bit-level cell M_i(j-1),configured to isolate the signal-storage state of the (j-1)-th bit-level cell M_i(j-1) in the (j-1)-th memory unit U_j-1 from the signal-storage state of the (j-2)-th bit-level cell M_i(j-2) in the (j-2)-th memory unit U_j-2, and to transfer a signal from the (j-2)-th bit-level cell M_i(j-2) to the (j-1)-th bit-level cell M_i(j-1) at the required timing determined by the clock signal, which is supplied through the clock signal supply line. In Fig. 13, because the input terminal of the (j-1)-th bit-level cell M_i(j-1) on the i-th row is connected to the inter-unit cell B_i(j-1), the signal charge stored in the (j-2)-th bit-level cell M_i(j-2) is fed to the second delay element D_i(j-1)2 through the inter-unit cell B_i(j-1) at the required timing, and the transfer operation of the signal charge is cut off thereafter.

In Fig. 13, although an example of the inter-unit cell B_i(j-1), which encompasses a single isolation transistor Q_i(j-1)3 having a first main-electrode connected to the output terminal of the (j-2)-th bit-level cell M_i(j-1), a second main-electrode connected to the input terminal of the (j-1)-th bit-level cell M_i(j-1) and a control electrode connected to the clock signal supply line, the structure of the inter-unit cell B_i(j-1) is not limited to the configuration illustrated in Fig. 13, and the inter-unit cell B_i(j-1) may be implemented by a clocked-circuit having a plurality of transistors, which can transfer the signal from the (j-2)-th bit-level cell M_i(j-2) to the (j-1)-th bit-level cell M_i(j-1) at the required timing determined by the clock signal.

Similar to the configuration of the j-th bit-level cell M_ij, the (j-1)-th bit-level cell M_i(j-1) encompasses a first nMOS transistor Q_i(j-1)1 having a drain electrode connected to the clock signal supply line through a first delay element D_i(j-1)1 and a gate electrode connected to the inter-unit cell B_i(j-1) through a second delay element D_i(j-1)2; a second nMOS transistor Q_i(j-1)2 having a drain electrode connected to the source electrode of the first nMOS transistor Q_i(j-1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i(j-1)configured to store the information of the bit-level cell M_i(j-1), connected in parallel with the second nMOS transistor Q_i(j-1)2 .

In the circuit configuration illustrated in Figs. 11 and 13, the second nMOS transistor Q_ij2 of the bit-level cell M_ij, serves as a reset-transistor configured to reset the signal charge stored in the capacitor C_ij, when the clock signal of high-level (or a logical level of "1") is applied to the gate electrode of the second nMOS transistor Q_ij2, discharging the signal charge already stored in the capacitor C_ij, and the second nMOS transistor Q_i(j-1)2 of the bit-level cell M_i(j-1) serves as a reset-transistor configured to reset the signal charge stored in the capacitor C_i(j-1), when the clock signal of high-level (or a logical level of "1") is applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, discharging the signal charge already stored in the capacitor C_i(j-1). Therefore, the isolation transistors Q_i(j-1)3 and Q_ij3 may be pMOS transistors, which can operate complementary with the second nMOS transistors Q_i(j-1)2 and Q_ij2, although Figs. 11 and 13 represent the transistor symbol of an nMOS transistor as the isolation transistors Q_i(j-1)3 and Q_ij3. That is, when the second nMOS transistors Q_i(j-1)2 and Q_ij2are conductive state for discharging the signal charge stored in the capacitors C_i(j-1)and C_ij, the isolation transistors Q_i(j-1)3 and Q_ij3 shall be cut-off state so as to establish the isolation between the memory units, and when the second nMOS transistors Q_i(j-1)2 and Q_ij2are cut-off state, the isolation transistors Q_i(j-1)3 and Q_ij3 shall be conductive state so as to transfer the signal charges between the memory units.

Alternatively, if the isolation transistors Q_i(j-1)3 and Q_ij3 are nMOS transistors, as the transistor symbol illustrates in Figs. 11 and 13, the isolation transistors Q_i(j-1)3 and Q_ij3 shall be high-speed transistors having a shorter rise time, a shorter period of conductive state, and a shorter fall time than the second nMOS transistors Q_i(j-1)2 and Q_ij2, which have larger stray capacitances and larger stray resistances associated with gate circuits and gate structures so that, when the second nMOS transistors Q_i(j-1)2 and Q_ij2are still in the cut-off state, the isolation transistors Q_i(j-1)3 and Q_ij3 becomes the conductive state very rapidly so as to transfer the signal charges between the memory units, and when the second nMOS transistors Q_i(j-1)2 and Q_ij2start slowly toward the conductive state for discharging the signal charge stored in the capacitors C_i(j-1)and C_ij, the isolation transistors Q_i(j-1)3 and Q_ij3 proceeds to become the cut-off state very rapidly so as to establish the isolation between the memory units. As a candidate for such high-speed transistors, a normally off type MOS static induction transistor (SIT) can be used, which represents triode-like I-V characteristic. N-channel MOSSIT can be considered as an extreme ultimate structure of the short channel nMOSFET. Owing to the triode-like I-V characteristic, because the on-state of the MOSSIT depends both on a gate voltage and a potential deference between the first and second main-electrodes, a very short time interval of the on-state can be achieved. Instead of the MOSSIT, any normally off type switching devices such as a tunneling SIT, which represent a very short on-state period like Dirac delta function, can be used.

Fig. 14(a) illustrates a timing diagram of a response of the bit-level cell M_i(j-1) illustrated in Fig. 13, and Fig. 14(b) illustrates a next timing diagram of a next response of the next bit-level cell M_ij illustrated in Fig. 13, to a waveform of a clock signal. In Figs. 14(a) and (b), the clock signal is supposed to swing periodically between the logical levels of "1" and "0" with the clock period TAU(Greek-letter) _clock, and the shaded rectangular area with backward diagonals illustrates a regime for a reset timing of the signal charges stored in the capacitors C_i(j-1) and C_ij, respectively, and further, the shaded rectangular area with forward diagonals illustrates a regime for a charge-transfer timing of the signal charges to the capacitors C_i(j-1) and C_ij, respectively.

That is, as shown in Fig. 14(a), if the signal charges stored in the capacitor C_i(j-1) is of the logical level of "1", although the first nMOS transistor Q_i(j-1)1 still keeps off-sate, the signal charge stored in the capacitor C_i(j-1) is being driven to be discharging, in the shaded rectangular area with backward diagonals. After the capacitor C_i(j-1) begins discharging, in the shaded rectangular area with forward diagonals, the first nMOS transistor Q_i(j-1)1 becomes active as a transfer-transistor, delayed by a predetermined delay time t_d1 determined by the first delay element D_i(j-1)1 implemented by the R-C delay circuit. And, when the signal stored in a previous bit-level cell M_i(j-2) is fed through the inter-unit cell B_i(j-1) to the gate electrode of the first nMOS transistor Q_i(j-1)1, the first nMOS transistor Q_i(j-1)1 transfers the signal stored in the previous bit-level cell M_i(j-2), further delayed by a predetermined delay time t_d2 determined by the second delay element D_i(j-1)2 to the capacitor C_i(j-1) in the shaded rectangular area with forward diagonals.

Similarly, as shown in Fig. 14(b), if the signal charges stored in the capacitor is of the logical level of "1", although the first nMOS transistor Q_ij1 still keeps off-sate, the signal charge stored in the capacitor C_ij is being driven to be discharging, in the shaded rectangular area with backward diagonals. After the capacitor C_ij begins discharging, in the shaded rectangular area with forward diagonals, the first nMOS transistor Q_ij1 becomes active as a transfer-transistor, delayed by a predetermined delay time t_d1 determined by the first delay element D_ij1 implemented by the R-C delay circuit. And, when the signal stored in a previous bit-level cell M_i(j-1) is fed through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 transfers the signal stored in the previous bit-level cell M_i(j-1), further delayed by a predetermined delay time t_d2 determined by the second delay element D_ij2 to the capacitor C_ij in the shaded rectangular area with forward diagonals

Fig. 15 illustrates a more detailed response of the bit-level cell M_i(j-1) illustrated in Fig. 13, which is one of the bit-level cells used in the computer system pertaining to the first embodiment of the present invention, to the waveform of the clock signal illustrated by thin solid line, for a case that both of the first delay element D_i(j-1)1 and the second delay element D_i(j-1)2 are implemented by R-C delay circuit, as illustrated in Fig. 12. The clock signal illustrated by thin solid line swings periodically between the logical levels of "1" and "0" with the clock period TAU_clock. In Fig. 15, time intervalTAU₁= TAU₂ = TAU₃ = TAU₄ is defined to be a quarter of the clock period TAU_clock(= TAU_clock /4).

In a normal operation of the marching memory, the signal charge stored in the capacitor C_i(j-1) is actually either of the logical level of "0" or"1", as illustrated in Figs. 16 (a)-(d). If the signal charge stored in the capacitor C_i(j-1) is of the logical level of "1", as illustrated in Figs. 16 (c) and (d), although the first nMOS transistor Q_i(j-1)1 still keeps off-sate, the capacitor C_i(j-1) can begin discharging at the beginning of the time intervalTAU₁, because the second nMOS transistor Q_i(j-1)2 becomes active when the clock signal of the high-level is applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, under the assumption that an ideal operation of the second nMOS transistor Q_i(j-1)2 with no delay can be approximated. Therefore, if the signal charge stored in the capacitor C_i(j-1) is actually of the logical level of "1", after the clock signal of high-level has been applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, as illustrated by the thin solid line in Fig. 15, and the signal charge stored in the capacitor C_i(j-1) will be discharged, and thereafter, the first nMOS transistor Q_i(j-1)1 becomes active as a transfer-transistor, delayed by a predetermined delay time t_d1 determined by the first delay element D_i(j-1)1 implemented by the R-C delay circuit. In Fig. 15, the change of the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 is illustrated by dash-dotted line.

And, as illustrated by a thick solid line in Fig. 15, when the signal level of "1" stored in a previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) on the i-th row through the inter-unit cell B_i(j-1) to the gate electrode of the first nMOS transistor Q_i(j-1)1, the first nMOS transistor Q_i(j-1)1 transfers the signal level of "1" stored in the previous bit-level cell M_i(j-2), further delayed by a predetermined delay time t_d2 determined by the second delay element D_i(j-1)2 to the capacitor C_i(j-1). Alternatively, as illustrated by a broken line in Fig. 15, when the signal level of "0" stored in a previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) to the gate electrode of the first nMOS transistor Q_i(j-1)1, the first nMOS transistor Q_i(j-1)1 transfers the signal level of "0" stored in the previous bit-level cell M_i(j-2), further delayed by the predetermined delay time t_d2 to the capacitor C_i(j-1). An output node N_out connecting the source electrode of the first nMOS transistor Q_i(j-1)1 and the drain electrode of the second nMOS transistor Q_i(j-1)2 serves as an output terminal of the bit-level cell M_i(j-1), and the output terminal delivers the signal stored in the capacitor C_i(j-1) to the next bit-level cell on the i-th row.

As illustrated by the thin solid line in Fig. 15, when the clock signal becomes the logical level of "1", the second nMOS transistor Q_i(j-1)2 begins to discharge the signal charge, which is already stored in the capacitor C_i(j-1) at a previous clock cycle. And, after the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_i(j-1) is completely discharged to the potential of the logical level of "0", the first nMOS transistor Q_i(j-1)1 becomes active as the transfer-transistor, delayed by the predetermined delay time t_d1 determined by the first delay element D_i(j-1)1. The delay time t_d1 may be set to be equal to 1/4TAU_clock= TAU₁ preferably.

Thereafter, when the signal stored in the previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) to the gate electrode of the first nMOS transistor Q_i(j-1)1 through the inter-unit cell B_i(j-1), as illustrated by thick solid line and broken line, the first nMOS transistor Q_i(j-1)1 transfers the signal stored in the previous bit-level cell M_i(j-2), further delayed by the predetermined delay time t_d2 determined by the second delay element D_i(j-1)2 implemented by the R-C delay circuit to the capacitor C_i(j-1).

For example, if the logical level of "1" stored in the previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) to the gate electrode of the first nMOS transistor Q_i(j-1)1 as illustrated by the thick solid line, the first nMOS transistor Q_i(j-1)1 becomes conductive state at the beginning of the time intervalTAU₃, and the logical level of "1" is stored in the capacitor C_i(j-1). On the other hand, if the logical level of "0" stored in the previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) to the gate electrode of the first nMOS transistor Q_i(j-1)1 as illustrated by the broken line, the first nMOS transistor Q_i(j-1)1 keeps the cut-off state, and the logical level of "0" is maintained in the capacitor C_i(j-1). Therefore, the bit-level cell M_i(j-1) can establish "a marching AND-gate" operation. The delay time t_d2 shall be longer than the delay time t_d1, and the delay time t_d2 may be set to be equal to 1/2TAU_clock preferably.

Because the clock signal swings periodically between the logical levels of "1" and "0", with the clock period TAU_clock as illustrated by the thin solid line, then, the clock signal becomes the logical level of "0" as time proceeds by 1/2TAU_clock, or at the beginning of the time intervalTAU₃, the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 begins to decay as illustrated by the dash-dotted line. If the inter-unit cell B_ij, inserted between the current bit-level cell M_i(j-1) and the next bit-level cell M_ij, is implemented by an nMOS transistor, the path between the output terminal of the current bit-level cell M_i(j-1) and the gate electrode of the first nMOS transistor Q_ij1 of the next bit-level cell M_ij, becomes the cut-off state by the logical level of "0" of the clock signal being applied to the gate electrode of the nMOS transistor, and therefore, the output node N_out connecting the source electrode of the first nMOS transistor Q_i(j-1)1 and the drain electrode of the second nMOS transistor Q_i(j-1)2 cannot deliver the signal transferred from the previous bit-level cell M_i(j-2) further to the next bit-level cell M_ij like duckpins in the time intervalsTAU₃ andTAU₄, and the signal is blocked to be domino transferred to the gate electrode of the next first nMOS transistor Q_ij1. Because the first nMOS transistor Q_i(j-1)1 becomes the cut-off state in the time intervalsTAU₃ andTAU₄, the potential at the output node N_out is kept in a floating state, and the signal states stored in the capacitor C_i(j-1) are held.

When the clock signal becomes the logical level of "1" again, as illustrated by the thin solid line in a next column of Fig. 15, the output node N_out connecting the source electrode of the first nMOS transistor Q_i(j-1)1 and the drain electrode of the second nMOS transistor Q_i(j-1)2, which is serving as the output terminal of the bit-level cell M_i(j-1), can deliver the signal stored in the capacitor C_i(j-1) to the next bit-level cell M_ij at the next clock cycle because the inter-unit cell B_ij becomes conductive state, and the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 increase as illustrated by the dash-dotted line.

Figs. 16 (a)-(d) illustrate four modes of signal-transferring operations, respectively, focusing to the bit-level cell M_ij illustrated in Figs. 11 and 13, the bit-level cell M_ij is one of the bit-level cells arrayed sequentially in the j-th memory unit U_j, the j-th memory unit U_j stores information of byte size or word size by the sequence of bit-level cells arrayed sequentially in the j-th memory unit U_j. In the computer system pertaining to the first embodiment of the present invention, the information of byte size or word size arrayed sequentially marches side by side from a previous memory unit to a next memory unit, pari passu. In Figs. 16 (a)-(d), the clock signal is supplied by the clock signal supply line CLOCK so as to swing periodically between the logical levels of "1" and "0" with the clock period TAU_clock, while the clock signal supply line CLOCK serves as a power supply line.

Figs. 16(a) and (b) illustrate the cases when the logical level of "0" is stored by previous clock signal into the capacitor C_ij, and Figs. 16 (c) and (d) illustrate the cases when the logical level of "1" is stored by previous clock signal into the capacitor C_ij as one of the signal in the information of byte size or word size. As illustrated in Fig. 16(a), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "0", if the signal of the logical level of "0", which is stored in a previous bit-level cell M_i(j-1), as one of the signal in the information of byte size or word size to be transferred in a cooperative way, is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij (the illustration is omitted) to the gate electrode of the first nMOS transistor Q_ij1, in the timing the signal charge stored in the capacitor C_ij keeping the logical level of "0", because the first nMOS transistor Q_ij1 keeps off-state, the output node N_out connecting the source electrode of the first nMOS transistor Q_ij1 and the drain electrode of the second nMOS transistor Q_ij2 delivers the signal level of "0", which is maintained in the capacitor C_ij, to the next bit-level cell on the i-th row, so as to execute marching AND-gate operation of 0 + 1= 0 with an input signal of "1" provided by the clock signal.

Similarly, as illustrated in Fig. 16(b), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "0", if the signal of the logical level of "1" stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, in the timing the signal charge stored in the capacitor C_ij keeps the logical level of "0", the first nMOS transistor Q_ij1 begins turning-on for transferring the signal of the logical level of "1" stored in the previous bit-level cell M_i(j-1) to the capacitor C_ij so that the logical level of "1" can be stored in the capacitor C_ij, and the output node N_out delivers the signal level of "1" stored in the capacitor C_ij to the next bit-level cell on the i-th row, so as to execute marching AND-gate operation 1 + 1= 1 with an input signal of "1" provided by the clock signal.

On the contrary, as illustrated in Fig. 16(c), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "1", if the signal of the logical level of "0", which is stored in a previous bit-level cell M_i(j-1), is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, after the timing when the signal charge stored in the capacitor C_ij is completely discharged to establish the logical level of "0", because the first nMOS transistor Q_ij1 keeps off-state, the output node N_out delivers the signal level of "0" stored in the capacitor C_ij to the next bit-level cell on the i-th row, so as to execute marching AND-gate operation of 0 + 1= 0 with an input signal of "1" provided by the clock signal.

Similarly, as illustrated in Fig. 16(d), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "1", if the signal of the logical level of "1" stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, after the timing when the signal charge stored in the capacitor C_ij is completely discharged to establish the logical level of "0", the first nMOS transistor Q_ij1 begins turning-on for transferring the signal of the logical level of "1" stored in the previous bit-level cell M_i(j-1) to the capacitor C_ij so that the logical level of "1" can be stored in the capacitor C_ij, and the output node N_out delivers the signal level of "1" stored in the capacitor C_ij to the next bit-level cell on the i-th row, so as to execute marching AND-gate operation 1 + 1= 1 with an input signal of "1" provided by the clock signal.

Similar to the configuration illustrated in Fig. 11, although an inter-unit cell B_ij is inserted between the (j-1)-th bit-level cell M_ij-1 and the j-th bit-level cell M_ij, and the j-th bit-level cell M_ij encompasses the first nMOS transistor Q_ij1 having the drain electrode connected to the clock signal supply line through the first delay element D_ij1 and the gate electrode connected to the inter-unit cell B_ij through the second delay element D_ij2; the second nMOS transistor Q_ij2 having the drain electrode connected to the source electrode of the first nMOS transistor Q_ij1, the gate electrode connected to the clock signal supply line, and the source electrode connected to the ground potential; and the capacitor C_ijconfigured to store the information of the bit-level cell M_ij, connected in parallel with the second nMOS transistor Q_ij2, the features such that the first delay element D_ij1 is implemented by a first diode D_1a, and the second delay element D_ij2 is implemented by a tandem connection of a second diode D_2a and a third diode D_3a is distinguishable from the configuration illustrated in Fig. 11.

Although any p-n junction diode can be represented by an equivalent circuit encompassing resistors including the series resistance such as the diffusion resistance, the lead resistance, the ohmic contact resistance and the spreading resistance, etc., and capacitors including the diode capacitance such as the junction capacitance or the diffusion capacitance, and a single diode or a tandem connection of diodes can serve as "resistive-capacitive delay" or "R-C delay", because the value of "R-C delay" can be made much smaller than the values achieved by the specialized and dedicated R-C elements such as the first meandering line 91 and the second meandering line 97 illustrated in Figs. 9 and 12, the operation of the j-th bit-level cell M_ij with the inter-unit cell B_ij illustrated in Fig. 17 can achieve more preferable operation than the operation achieved by the configuration illustrated in Fig. 12. That is, the operation of the j-th bit-level cell M_ij with the inter-unit cell B_ij illustrated in Fig. 17 can approaches to an ideal delay performance illustrated in Figs. 7A and 7B, in which any rise time and fall time are not illustrated, and wave forms of the pulses are illustrated by ideal rectangular shape. In addition to the performance by the configuration illustrated in Figs. 11 and 12, because the tandem connection of the second diode D_2a and the third diode D_3a can block efficiently the flow of the reverse-directional current, the configuration implemented by a combination of the j-th bit-level cell M_ij with the inter-unit cell B_ij illustrated in Fig. 17 can achieve a better isolation between the signal-storage state of the (j-1)-th bit-level cell M_i(j-1) and the signal-storage state of the j-th bit-level cell M_ij, even if the signal of the lower logical level of "0" stored in the previous bit-level cell M_i(j-1) is fed to the gate electrode of the first nMOS transistor Q_ij1 through the inter-unit cell B_ij.

In Fig. 18, in addition to the configuration illustrated in Fig. 17, another inter-unit cell B_i(j-1) is provided between the (j-2)-th bit-level cell M_i(j-2) and the (j-1)-th bit-level cell M_i(j-1),configured to isolate the signal-storage state of the (j-1)-th bit-level cell M_i(j-1) in the (j-1)-th memory unit U_j-1 from the signal-storage state of the (j-2)-th bit-level cell M_i(j-2) in the (j-2)-th memory unit U_j-2, and to transfer a signal from the (j-2)-th bit-level cell M_i(j-2) to the (j-1)-th bit-level cell M_i(j-1) at the required timing determined by the clock signal, which is supplied through the clock signal supply line. In Fig. 18, because the input terminal of the (j-1)-th bit-level cell M_i(j-1) is connected to the inter-unit cell B_i(j-1), the signal charge stored in the (j-2)-th bit-level cell M_i(j-2) is fed to the second delay element D_i(j-1)2 through the inter-unit cell B_i(j-1) at the required timing, and the transfer of the signal charge is cut off thereafter.

Similar to the configuration of the j-th bit-level cell M_ij, the (j-1)-th bit-level cell M_i(j-1) encompasses a first nMOS transistor Q_i(j-1)1 having a drain electrode connected to the clock signal supply line through a first delay element D_i(j-1)1 and a gate electrode connected to the inter-unit cell B_i(j-1) through a second delay element D_i(j-1)2; a second nMOS transistor Q_i(j-1)2 having a drain electrode connected to the source electrode of the first nMOS transistor Q_i(j-1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i(j-1)configured to store the information of the bit-level cell M_i(j-1), connected in parallel with the second nMOS transistor Q_i(j-1)2. Here, the first delay element D_i(i-1)1 is implemented by a first diode D_1b, and the second delay element D_i(i-1)2 is implemented by a tandem connection of a second diode D_2b and a third diode D_3b.

As explained above, because a single diode or a tandem connection of diodes can serve as "resistive-capacitive delay" or "R-C delay", the operation of the (j-1)-th bit-level cell M_i(i-1) with the inter-unit cell B_i(i-1) illustrated in Fig. 18 is substantially same as the operation of the configuration illustrated in Fig. 13. In addition to the performance by the configuration illustrated in Fig. 13, because the tandem connection of the second diode D_2b and the third diode D_3b can block efficiently the flow of the reverse-directional current, the configuration implemented by a combination of the (j-1)-th bit-level cell M_i(i-1) with the inter-unit cell B_i(i-1) illustrated in Fig. 18 can achieve a better isolation between the signal-storage state of the (j-2)-th bit-level cell M_i(j-2) and the signal-storage state of the (j-1)-th bit-level cell M_i(i-1), even if the signal of the lower logical level of "0" stored in the previous bit-level cell M_i(j-2) is fed to the gate electrode of the first nMOS transistor Q_i(i-1)1 through the inter-unit cell B_i(i-1).

In actual semiconductor devices, because there are many parasitic resistances (stray resistances) and many parasitic capacitances (stray capacitances) associated with wirings, gate structures, electrode structures, and junction structures are inherent, in a very high speed operation of the marching main memory, the delineation of extrinsic resistor elements and capacitor elements can be omitted, if the parasitic resistances and the parasitic capacitances can achieve the required delay times t_d1 , t_d2 compared with operation speed of the marching main memory. Therefore, in the configuration illustrated in Figs. 11-13 and 16, the first delay elements D_i(j-1)1 and D_ij1 can be omitted, as illustrated in Figs. 19, 20 and 22.

In one of other examples of the bit-level cells used in the computer system pertaining to the first embodiment of the present invention illustrated in Fig. 19, although the j-th bit-level cell M_ij encompasses a first nMOS transistor Q_ij1, similar to the configuration illustrated in Fig. 11, the first nMOS transistor Q_ij1 has a drain electrode directly connected to the clock signal supply line, and the first delay element D_ij1 employed in the configuration illustrated in Fig. 11 is omitted. The feature that the first nMOS transistor Q_ij1 has a gate electrode connected to the inter-unit cell B_ij through a signal-delay element D_ij, which corresponds to the second delay element D_ij2 illustrated in Fig. 11, and the second nMOS transistor Q_ij2 has a drain electrode connected to a source electrode of the first nMOS transistor Q_ij1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential, and a capacitor C_ijconfigured to store the information of the bit-level cell M_ij, connected in parallel with the second nMOS transistor Q_ij2 is substantially same as the configuration illustrated in Fig. 11.

In the other example of the bit-level cell pertaining to the first embodiment illustrated in Fig. 19, similar to the configuration illustrated in Figs. 11-13 and 16, the inter-unit cell B_ij is further provided so as to isolate the signal-storage state of the j-th bit-level cell M_ij in the j-th memory unit U_j from the signal-storage state of the (j-1)-th bit-level cell M_ij-1 in the (j-1)-th memory unit U_j-1. Furthermore, the inter-unit cell B_ij transfers a signal from the (j-1)-th bit-level cell M_ij-1 to the j-th bit-level cell M_ij at a required timing determined by a clock signal, which is supplied through the clock signal supply line. Because the j-th memory unit U_j stores information of byte size or word size by the sequence of bit-level cells arrayed in the j-th memory unit U_j, and the (j-1)-th memory unit U_j-1 stores information of byte size or word size by the sequence of bit-level cells arrayed in the (j-1)-th memory unit U_j-1, a sequence of inter-unit cells arrayed in parallel with the memory units U_j-1 and U_j transfers the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line so that the information of byte size or word size can march along a predetermined direction, pari passu.

As illustrated in Fig. 19, because the input terminal of the j-th bit-level cell M_ij on the i-th row is connected to the inter-unit cell B_ij, the signal charge stored in the (j-1)-th bit-level cell M_ij-1 is fed to the signal-delay element D_ij through the inter-unit cell B_ij at the required timing, and the transfer operation of the signal charge is cut off at periods other than the required timing.

In Fig. 20, in addition to the configuration illustrated in Fig. 19, another inter-unit cell B_i(j-1) is provided between the (j-2)-th bit-level cell M_i(j-2) and the (j-1)-th bit-level cell M_i(j-1),configured to isolate the signal-storage state of the (j-1)-th bit-level cell M_i(j-1) in the (j-1)-th memory unit U_j-1 from the signal-storage state of the (j-2)-th bit-level cell M_i(j-2) in the (j-2)-th memory unit U_j-2, and to transfer a signal from the (j-2)-th bit-level cell M_i(j-2) to the (j-1)-th bit-level cell M_i(j-1) at the required timing determined by the clock signal, which is supplied through the clock signal supply line. In Fig. 20, because the input terminal of the (j-1)-th bit-level cell M_i(j-1) on the i-th row is connected to the inter-unit cell B_i(j-1), the signal charge stored in the (j-2)-th bit-level cell M_i(j-2) is fed to the signal-delay element D_i(j-1) through the inter-unit cell B_i(j-1) at the required timing, and the transfer operation of the signal charge is cut off thereafter.

Similar to the configuration of the j-th bit-level cell M_ij, the (j-1)-th bit-level cell M_i(j-1) encompasses a first nMOS transistor Q_i(j-1)1 having a drain electrode directly connected to the clock signal supply line and a gate electrode connected to the inter-unit cell B_i(j-1) through a signal-delay element D_i(j-1); a second nMOS transistor Q_i(j-1)2 having a drain electrode connected to the source electrode of the first nMOS transistor Q_i(j-1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i(j-1)configured to store the information of the bit-level cell M_i(j-1), connected in parallel with the second nMOS transistor Q_i(j-1)2 .

In the circuit configuration, as one of other examples of the bit-level cells pertaining to the first embodiment, illustrated in Figs. 19 and 20, the second nMOS transistor Q_ij2 of the bit-level cell M_ij, serves as a reset-transistor configured to reset the signal charge stored in the capacitor C_ij, when the clock signal of high-level (or a logical level of "1") is applied to the gate electrode of the second nMOS transistor Q_ij2, discharging the signal charge already stored in the capacitor C_ij, and the second nMOS transistor Q_i(j-1)2 of the bit-level cell M_i(j-1) serves as a reset-transistor configured to reset the signal charge stored in the capacitor C_i(j-1), when the clock signal of high-level (or a logical level of "1") is applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, discharging the signal charge already stored in the capacitor C_i(j-1).

In Figs. 19 and 20, the isolation transistors Q_i(j-1)3 and Q_ij3 shall be high-speed transistors having a shorter rise time, a shorter period of conductive state, and a shorter fall time than the second nMOS transistors Q_i(j-1)2 and Q_ij2, which have larger stray capacitances and larger stray resistances associated with gate circuits and gate structures so that, when the second nMOS transistors Q_i(j-1)2 and Q_ij2are still in the cut-off state, the isolation transistors Q_i(j-1)3 and Q_ij3 becomes the conductive state very rapidly so as to transfer the signal charges between the memory units, and when the second nMOS transistors Q_i(j-1)2 and Q_ij2start slowly toward the conductive state for discharging the signal charge stored in the capacitors C_i(j-1)and C_ij, the isolation transistors Q_i(j-1)3 and Q_ij3 proceeds to become the cut-off state very rapidly so as to establish the isolation between the memory units.

Fig. 21 illustrates a detailed response of the bit-level cell M_i(j-1) illustrated in Fig. 20, which is one of other examples of the bit-level cells used in the computer system pertaining to the first embodiment of the present invention, to the waveform of the clock signal illustrated by thin solid line, for a case that the signal-delay element D_i(j-1) is implemented by R-C delay circuit. The clock signal illustrated by thin solid line swings periodically between the logical levels of "1" and "0" with the clock period TAU_clock. In Fig. 21, time intervalTAU₁= TAU₂ = TAU₃ = TAU₄ is defined to be a quarter of the clock period TAU_clock(= TAU_clock /4).

In a normal operation of the marching memory, the signal charge stored in the capacitor C_i(j-1) is actually either of the logical level of "0" or"1", as illustrated in Figs. 22 (a)-(d). If the signal charge stored in the capacitor C_i(j-1) is of the logical level of "1", as illustrated in Figs. 22 (c) and (d), although the first nMOS transistor Q_i(j-1)1 still keeps off-sate because the potential of the gate electrode of the first nMOS transistor Q_i(j-1)1 is delayed by the signal-delay element D_i(j-1), the capacitor C_i(j-1) can begin discharging at the beginning of the time intervalTAU₁, because the second nMOS transistor Q_i(j-1)2 becomes active rapidly when the clock signal of the high-level is applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, under the assumption that an ideal operation of the second nMOS transistor Q_i(j-1)2 with no delay can be approximated. Therefore, if the signal charge stored in the capacitor C_i(j-1) is actually of the logical level of "1", after the clock signal of high-level has been applied to the gate electrode of the second nMOS transistor Q_i(j-1)2, as illustrated by the thin solid line in Fig. 21, and the signal charge stored in the capacitor C_i(j-1) will be discharged to the logical level of "0", and at the same time approximately, the first nMOS transistor Q_i(j-1)1 is prepared to be active as a transfer-transistor, delayed by a negligibly-short delay time determined by parasitic elements implemented by stray resistance and stray capacitance. In Fig. 21, the change of the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 is illustrated exaggeratingly by dash-dotted line.

And, as illustrated by a thick solid line in Fig. 21, when the signal level of "1" stored in a previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) through the inter-unit cell B_i(j-1) to the gate electrode of the first nMOS transistor Q_i(j-1)1, the first nMOS transistor Q_i(j-1)1 turns on, and the first nMOS transistor Q_i(j-1)1 transfers the signal level of "1" stored in the previous bit-level cell M_i(j-2), delayed by a predetermined delay time t_d2 determined by the signal-delay element D_i(j-1) to the capacitor C_i(j-1). Alternatively, as illustrated by a broken line in Fig. 21, when the signal level of "0" stored in a previous bit-level cell M_i(j-2) is fed from the previous bit-level cell M_i(j-2) to the gate electrode of the first nMOS transistor Q_i(j-1)1, the first nMOS transistor Q_i(j-1)1 keeps off-state. At this instant of time, because the capacitor C_i(j-1) still keep the logical level of "0", the first nMOS transistor Q_i(j-1)1 transfers equivalently the signal level of "0" stored in the previous bit-level cell M_i(j-2). An output node N_out serving as an output terminal of the bit-level cell M_i(j-1) delivers the signal stored in the capacitor C_i(j-1) to the next bit-level cell on the i-th row.

Because the clock signal swings periodically between the logical levels of "1" and "0", with the clock period TAU_clock as illustrated by the thin solid line, then, the clock signal becomes the logical level of "0" as time proceeds by 1/2TAU_clock, or at the beginning of the time intervalTAU₃, the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 begins to decay rapidly as illustrated exaggeratingly by the dash-dotted line. If the inter-unit cell B_ij, inserted between the current bit-level cell M_i(j-1) and the next bit-level cell M_ij, is implemented by an nMOS transistor, the path between the output terminal of the current bit-level cell M_i(j-1) and the gate electrode of the first nMOS transistor Q_ij1 of the next bit-level cell M_ij, becomes the cut-off state by the logical level of "0" of the clock signal being applied to the gate electrode of the nMOS transistor, and therefore, the output node N_out cannot deliver the signal transferred from the previous bit-level cell M_i(j-2) further to the next bit-level cell M_ij like duckpins in the time intervalsTAU₃ andTAU₄, and the signal is blocked to be domino transferred to the gate electrode of the next first nMOS transistor Q_ij1. Because the first nMOS transistor Q_i(j-1)1 becomes the cut-off state in the time intervalsTAU₃ andTAU₄, the potential at the output node N_out is kept in a floating state, and the signal states stored in the capacitor C_i(j-1) are held.

When the clock signal becomes the logical level of "1" again, as illustrated by the thin solid line in a next column of Fig. 21, the output node N_out connecting the source electrode of the first nMOS transistor Q_i(j-1)1 and the drain electrode of the second nMOS transistor Q_i(j-1)2, which is serving as the output terminal of the bit-level cell M_i(j-1), can deliver the signal stored in the capacitor C_i(j-1) to the next bit-level cell M_ij at the next clock cycle because the inter-unit cell B_ij becomes conductive state, and the potential at the drain electrode of the first nMOS transistor Q_i(j-1)1 increase as illustrated exaggeratingly by the dash-dotted line.

Figs. 22 (a)-(d) illustrate four modes of signal-transferring operations, respectively, focusing to the bit-level cell M_ij illustrated in Figs. 19 and 20, the bit-level cell M_ij is one of the bit-level cells arrayed sequentially in the j-th memory unit U_j, the j-th memory unit U_j stores information of byte size or word size by the sequence of bit-level cells arrayed sequentially in the j-th memory unit U_j. In the computer system pertaining to the first embodiment of the present invention, the information of byte size or word size arrayed sequentially marches side by side from a previous memory unit to a next memory unit, pari passu. In Figs. 22 (a)-(d), the clock signal is supplied by the clock signal supply line CLOCK so as to swing periodically between the logical levels of "1" and "0" with the clock period TAU_clock, while the clock signal supply line CLOCK serves as a power supply line.

Figs. 22(a) and (b) illustrate the cases when the logical level of "0" is stored by previous clock signal into the capacitor C_ij, and Figs. 22 (c) and (d) illustrate the cases when the logical level of "1" is stored by previous clock signal into the capacitor C_ij as one of the signal in the information of byte size or word size. As illustrated in Fig. 22(a), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "0", if the signal of the logical level of "0", which is stored in a previous bit-level cell M_i(j-1), as one of the signal in the information of byte size or word size to be transferred in a cooperative way, is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij (the illustration is omitted) to the gate electrode of the first nMOS transistor Q_ij1, the first nMOS transistor Q_ij1 keeps off-state. At this instant of time, because the capacitor C_ij still keep the logical level of "0", the first nMOS transistor Q_i(j-1)1 transfers equivalently the logical level of "0" to the capacitor C_ij. Then, the output node N_out delivers the signal level of "0", which is maintained in the capacitor C_ij, to the next bit-level cell as illustrated in Fig. 22(a).

Similarly, as illustrated in Fig. 22(b), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "0", if the signal of the logical level of "1" stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, in the timing the signal charge stored in the capacitor C_ij keeps the logical level of "0", the first nMOS transistor Q_ij1 begins turning-on for transferring the signal of the logical level of "1" stored in the previous bit-level cell M_i(j-1) to the capacitor C_ij so that the logical level of "1" can be stored in the capacitor C_ij, and the output node N_out delivers the signal level of "1" stored in the capacitor C_ij to the next bit-level cell as illustrated in Fig. 22(b).

On the contrary, as illustrated in Fig. 22(c), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "1", if the signal of the logical level of "0", which is stored in a previous bit-level cell M_i(j-1), is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, after the timing when the signal charge stored in the capacitor C_ij is completely discharged to establish the logical level of "0", the first nMOS transistor Q_ij1 keeps off-state. Then, the output node N_out delivers the signal level of "0" stored in the capacitor C_ij to the next bit-level cell as illustrated in Fig. 22(c).

Similarly, as illustrated in Fig. 22(d), in a case when the signal charge previously stored in the capacitor C_ij is of the logical level of "1", if the signal of the logical level of "1" stored in a previous bit-level cell M_i(j-1) is fed from the previous bit-level cell M_i(j-1) through the inter-unit cell B_ij to the gate electrode of the first nMOS transistor Q_ij1, after the timing when the signal charge stored in the capacitor C_ij is completely discharged to establish the logical level of "0", the first nMOS transistor Q_ij1 turns on, and the first nMOS transistor Q_ij1 transfers the signal of the logical level of "1" stored in the previous bit-level cell M_i(j-1) to the capacitor C_ij. Then, the output node N_out delivers the signal level of "1" stored in the capacitor C_ij to the next bit-level cell as illustrated in Fig. 22(d).

As above-mentioned, with an input signal of "1" provided by the clock signal and another input signal of "1" or "0" provided by the previous bit-level cell M_i(j-1), the bit-level cell M_ij can establish "a marching AND-gate" operations of:

1 + 1 = 1
1 + 0 = 1,

and with an input signal of "0" provided by the clock signal and another input signal of "1" or "0" provided by the previous bit-level cell M_i(j-1), the bit-level cell M_ij can establish "the marching AND-gate" operations of:

0 + 1 = 0
0 + 0 = 0.

Therefore, in a gate-level representation of the cell array corresponding to the marching main memory 31 illustrated in Fig. 4, as illustrated in Fig. 23, a first cell M₁₁ allocated at the leftmost side on a first row and connected to an input terminal I₁ encompasses a capacitor C₁₁configured to store the information, and a marching AND-gate G₁₁having one input terminal connected to the capacitor C₁₁, the other input configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G₂₁ assigned to the adjacent second cell M₂₁ on the first row. An example of the response to the waveform of the clock signal is illustrated in Fig. 7C. When the logical values of "1" of the clock signal is fed to the other input terminal of the marching AND-gate G₁₁, the information stored in the capacitor C₁₁ is transferred to a capacitor C₁₂, assigned to the adjacent second cell M₁₂, and the capacitor C₁₂stores the information. Namely, the second cell M₁₂ on the first row of the gate-level representation of cell array implementing the marching main memory 31 encompasses the capacitor C₁₂and a marching AND-gate G₁₂, which has one input terminal connected to the capacitor C₁₂, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G₁₃assigned to the adjacent third cell M₁₃ on the first row. Similarly the third cell M₁₃ on the first row of the gate-level representation of cell array implementing the marching main memory 31 encompasses a capacitor C₁₃configured to store the information, and a marching AND-gate G₁₃having one input terminal connected to the capacitor C₁₃, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate assigned to the adjacent fourth cell, although the illustration of the fourth cellis omitted. Therefore, when the logical values of "1" is fed to the other input terminal of the marching AND-gate G₁₂, the information stored in the capacitor C₁₂ is transferred to the capacitor C₁₃, assigned to the third cell M₁₃, and the capacitor C₁₃stores the information, and when the logical values of "1" is fed to the other input terminal of the marching AND-gate G₁₃, the information stored in the capacitor C₁₃ is transferred to the capacitor, assigned to the fourth cell. Furthermore, a (n-1)-th cell M_{1, n-1} on the first row of the gate-level representation of cell array implementing the marching main memory 31 encompasses a capacitor C_{1, n-1}configured to store the information, and a marching AND-gate G_{1, n-1}having one input terminal connected to the capacitor C_{1, n-1}, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G_1n assigned to the adjacent n-th cell M_1n, which is allocated at the rightmost side on the first row and connected to an output terminal O₁. Therefore, each of the cells M₁₁, M₁₂, M₁₃,........., M_{1, n-1}, M_1n stores the information, and transfers the information synchronously with the clock signal, step by step, toward the output terminals O₁, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

Similarly, in a gate-level representation of cell array implementing the marching main memory 31 illustrated in Fig. 23, a first cell M₂₁ allocated at the leftmost side on a second row and connected to an input terminal I₂ encompasses a capacitor C₂₁, and a marching AND-gate G₂₁having one input terminal connected to the capacitor C₂₁, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G₂₁ assigned to the adjacent second cell M₂₁ on the second row. The second cell M₂₂ on the second row of the gate-level representation of cell array implementing the marching main memory 31 encompasses the capacitor C₂₂and a marching AND-gate G₂₂, which has one input terminal connected to the capacitor C₂₂, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G₂₃assigned to the adjacent third cell M₂₃ on the second row. Similarly the third cell M₂₃ on the second row of the gate-level representation of cell array implementing the marching main memory 31 encompasses a capacitor C_23, and a marching AND-gate G₂₃having one input terminal connected to the capacitor C₂₃, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate assigned to the adjacent fourth cell. Furthermore, a (n-1)-th cell M_{2, n-1} on the second row of the gate-level representation of cell array implementing the marching main memory 31 encompasses a capacitor C_{2, n-1,}and a marching AND-gate G_{2, n-1}having one input terminal connected to the capacitor C_{2, n-1}, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the next marching AND-gate G_1n assigned to the adjacent n-th cell M_1n, which is allocated at the rightmost side on the second row and connected to an output terminal O₁. Therefore, each of the cells M₂₁, M₂₂, M₂₃,........., M_{2, n-1}, M_1n on the second row stores the information, and transfers the information synchronously with the clock signal, step by step, toward the output terminals O₁, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

On a third row, a first cell M₃₁ allocated at the leftmost side and connected to an input terminal I₃, a second cell M₃₂ adjacent to the first cell M₃₁, a third cell M₃₃ adjacent to the second cell M₃₂, ........., a (n-1)-th cell M_{3, n-1}, and an n-th cell M_3n, which is allocated at the rightmost side on the third row and connected to an output terminal O₃ are aligned. And, each of the cells M₃₁, M₃₂, M₃₃,........., M_{3, n-1}, M_3n on the third row stores the information, and transfers the information synchronously with the clock signal, step by step, toward the output terminals O₃, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

On a (m-1)-th row, a first cell M_(m-1),1 allocated at the leftmost side and connected to an input terminal I_m-1,, a second cell M_(m-1),2 adjacent to the first cell M_(m-1),1, a third cell M_(m-1),3 adjacent to the second cell M_(m-1),2, ........., a (n-1)-th cell M_{(m-1), n-1}, and an n-th cell M_(m-1),n, which is allocated at the rightmost side on the (m-1)-th row and connected to an output terminal O_m-1, are aligned. And, each of the cells M_(m-1),1, M_(m-1),2, M_(m-1),3,........., M_{(m-1), n-1}, M_(m-1),n on the (m-1)-th row stores the information, and transfers the information synchronously with the clock signal, step by step, toward the output terminals O_m-1,, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

On a m-th row, a first cell M_m1 allocated at the leftmost side and connected to an input terminal I_m-1, a second cell M_m2 adjacent to the first cell M_m1, a third cell M_m3 adjacent to the second cell M_m2,........, a (n-1)-th cell M_m(n-1), and an n-th cell M_mn, which is allocated at the rightmost side on the m-th row and connected to an output terminal O_m, are aligned. And, each of the cells M_m1, M_m2, M_m3,........., M_m(n-1), M_mn on the m-th row stores the information, and transfers the information synchronously with the clock signal, step by step, toward the output terminals O_m, so as to provide the processor 11 with the stored information actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

Although one of the examples of the transistor-level configurations of the marching AND-gate G_ij is illustrated in Fig. 6, there are various circuit configurations to implement the marching AND-gate, which can be applied to the cell array implementing the marching main memory 31 in the computer system pertaining to the first embodiment of the present invention. Another example of the marching AND-gate G_ij, which can be applied to the cell array implementing the marching main memory 31, may be a configuration encompassing a CMOS NAND gate and a CMOS inverter connected to the output terminal of the CMOS NAND gate. Because the CMOS NAND gate requires two nMOS transistors and two pMOS transistors, and the CMOS inverter requires one nMOS transistor and one pMOS transistor, the configuration encompassing the CMOS NAND gate and the CMOS inverter requires six transistors. Furthermore, the marching AND-gate G_ij can be implemented by other circuit configurations such as resistor-transistor logics, or by various semiconductor elements, magnetic elements, superconductor elements, or single quantum elements, etc. which has a function of AND logic.

As illustrated in Fig. 23, the gate-level representation of cell array implementing the marching main memory 31 is as simple as the configuration of DRAM, where each of the bit-level cells M_ij (i = 1 to m; j = 1 to n) is represented by one capacitor and one marching AND-gate. Each of the vertical sequence of marching AND-gates G₁₁, G₂₁, G₃₁, ........, G_m-1,1, G_m1implementing the first memory unit U₁ shifts the sequence of signals from input terminals I₁, I₂, I₃,........., I_n-1, I_n to right along row-direction, or horizontal direction, based on clocks as illustrated in Fig. 7C. And, each of the vertical sequence of marching AND-gates G₁₂, G₂₂, G₃₂, ........, G_m-1,2, G_m2implementing the second memory unit U₂ shifts the sequence of signals of word size from left to right along row-direction based on clocks, each of the vertical sequence of marching AND-gates G₁₃, G₂₃, G₃₃, ........, G_m-1,3, G_m3implementing the third memory unit U₃ shifts the sequence of signals of word size from left to right along row-direction based on clocks, ............,each of the vertical sequence of marching AND-gates G_1,n-1, G_2,n-1, G_3,n-1, ........, G_m-1,n-1, G_m,n-1implementing the (n-1)-th memory unit U_n-1 shifts the sequence of signals of word size from left to right along row-direction based on clocks, and each of the vertical sequence of marching AND-gates G_1,n, G_2,n, G_3,n, ........, G_m-1,n, G_m,nimplementing the n-th memory unit U_n shifts the sequence of signals of word size from left to right to the output terminals O₁, O₂, O₃,........., O_n-1, O_nbased on clocks as illustrated in Fig. 7C. Especially, the time delay t_d1 , t_d2 in each of marching AND-gate G_ij (i = 1 to m; j = 1 to n) is significant to correctly perform the marching-shift actions in every memory units in the marching main memory 31 successively.

(REVERSE-DIRECTIONAL MARCHING MAIN MEMORY)
Although Figs. 3-23 illustrate the marching main memory which stores the information in each of memory units U₁, U₂, U₃,........., U_n-1, U_n and transfers the information synchronously with the clock signal, step by step, from input terminal toward the output terminal, Fig. 24 illustrates another marching main memory.

In Fig. 24, each of the memory units U₁, U₂, U₃,........., U_n-1, U_n stores the information including word size of data or instructions, and transfers in the reverse direction the information synchronously with the clock signal, step by step, toward the output terminals, provided from the processor 11 with the resultant data executed in the ALU 112.

Fig. 25(a) illustrates an array of i-th row of the m * n matrix (here, "m" is an integer determined by word size) in a cell-level representation of the another marching main memory illustrated in Fig. 24, which stores the information of bit level in each of cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,n and transfers the information synchronously with the clock signal, step by step in the reverse direction to the marching main memory illustrated in Figs. 3-23, namely from the output terminal OUT toward the input terminal IN.

As illustrated in Fig. 25(a), in a reverse-directional marching main memory, a bit-level cell M_in of the n-th column and on the i-th row, allocated at the rightmost side on the i-th row and connected to an input terminal IN encompasses a first nMOS transistor Q_in1 having a drain electrode connected to a clock signal supply line through a first delay element D_in1 and a gate electrode connected to the input terminal IN through a second delay element D_in2; a second nMOS transistor Q_in2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_in1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_inconfigured to store the information of the bit-level cell M_in, connected in parallel with the second nMOS transistor Q_in2, wherein an output node connecting the source electrode of the first nMOS transistor Q_in1 and the drain electrode of the second nMOS transistor Q_in2 serves as an output terminal of the bit-level cell M_in, configured to transfer the signal stored in the capacitor C_in to the next bit-level cell M_i2.

As illustrated in Fig. 25(b), the clock signal swings periodically between the logical levels of "1" and "0", with a predetermined clock period TAU_clock, and when the clock signal becomes the logical level of "1", the second nMOS transistor Q_in2 begins to discharge the signal charge, which is already stored in the capacitor C_in at a previous clock cycle. And, after the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_in is completely discharged to becomes the logical level of "0", the first nMOS transistor Q_in1 becomes active as the transfer transistor, delayed by the predetermined delay time t_d1 determined by the first delay element D_in1. The delay time t_d1 may be set to be equal to 1/4TAU_clock preferably. Thereafter, when the signal is fed from the input terminal IN to the gate electrode of the first nMOS transistor Q_in1, the first nMOS transistor Q_in1 transfers the signal stored in the previous bit-level cell M_i2, further delayed by the predetermined delay time t_d2 determined by the second delay element D_in2 to the capacitor C_in. For example, if the logical level of "1" is fed from the input terminal IN to the gate electrode of the first nMOS transistor Q_in1, the first nMOS transistor Q_in1 becomes conductive state, and the logical level of "1" is stored in the capacitor C_in. On the other hand, if the logical level of "0" is fed from the input terminal IN to the gate electrode of the first nMOS transistor Q_in1, the first nMOS transistor Q_in1 keeps cut-off state, and the logical level of "0" is maintained in the capacitor C_in. Therefore, the bit-level cell M_in can establish "a marching AND-gate" operation. The delay time t_d2 shall be longer than the delay time t_d1, and the delay time t_d2 may be set to be equal to 1/2TAU_clock preferably. When the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrodes of the first nMOS transistor Q_in1and the drain electrodes of the second nMOS transistor Q_in2 cannot deliver the signals, which are entered to the gate electrodes of the first nMOS transistor Q_in1, further to the next bit-level cell M_i2, at a time when time proceeds 1/2TAU_clock, as the signals is blocked to be transferred to the gate electrodes of the next first nMOS transistor Q_i21 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D_i22. And, at a time when time proceeds TAU_clock, when the next clock signal becomes the logical level of "1" again, a sequence of the second nMOS transistors

As illustrated in Fig. 25(a), in a reverse-directional marching main memory, a bit-level cell M_i(n-1) of the (n-1)-th column and on the i-th row, allocated at the second right side on the i-th row, encompasses a first nMOS transistor Q_i(n-1)1 having a drain electrode connected to the clock signal supply line through a first delay element D_i(n-1)1 and a gate electrode connected to the output terminal of the bit-level cell M_in through a second delay element D_i(n-1)2; a second nMOS transistor Q_i(n-1)2 having a drain electrode connected to a source electrode of the first nMOS transistor Q_i(n-1)1, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i(n-1)configured to store the information of the bit-level cell M_i(n-1), connected in parallel with the second nMOS transistor Q_i(n-1)2. When the clock signal becomes the logical level of "1", the second nMOS transistor Q_i(n-1)2 begins to discharge the signal charge, which is already stored in the capacitor C_i(n-1) at a previous clock cycle. And, as illustrated in Fig. 25(b), and the logical values of "1" is kept from time "t" to time "t+1" in the capacitor C_i(n-1). After the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_i(n-1) is completely discharged to becomes the logical level of "0", the first nMOS transistor Q_i(n-1)1 becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first delay element D_i(n-1)1. Thereafter, when the signal is fed from the output terminal of the bit-level cell M_in to the gate electrode of the first nMOS transistor Q_i(n-1)1, the first nMOS transistor Q_i(n-1)1 transfers the signal stored in the previous bit-level cell M_in, further delayed by the delay time t_d2 determined by the second delay element D_i(n-1)2 to the capacitor C_i(n-1). When the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrodes of the first nMOS transistor Q_i(n-1)1and the drain electrodes of the second nMOS transistor Q_i(n-1)2 cannot deliver the signals, which are entered to the gate electrodes of the first nMOS transistor Q_i(n-1)1, further to the next bit-level cell M_i(n-2), at a time when time proceeds 1/2TAU_clock, as the signals is blocked to be transferred to the gate electrodes of the next first nMOS transistor Q_i(n-2)1 (illustration is omitted) delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D_i(n-2)2(illustration is omitted).

Similarly the third cell M_i3 from the left, on the i-th row, of the reverse-directional marching main memory encompasses a first nMOS transistor Q_i31 having a drain electrode connected to the clock signal supply line through a first delay element D_i31 and a gate electrode connected to the output terminal of the bit-level cell M_i4(illustration is omitted) through a second delay element D_i32; a second nMOS transistor Q_i32 having a drain electrode connected to a source electrode of the first nMOS transistor Q_i31, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i3configured to store the information of the bit-level cell M_i3, connected in parallel with the second nMOS transistor Q_i32. When the clock signal becomes the logical level of "1", the second nMOS transistor Q_i32 begins to discharge the signal charge, which is already stored in the capacitor C_i3 at a previous clock cycle. After the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_i3 is completely discharged to becomes the logical level of "0", the first nMOS transistor Q_i31 becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first delay element D_i31. Thereafter, when the signal is fed from the output terminal of the bit-level cell M_i4 to the gate electrode of the first nMOS transistor Q_i31, the first nMOS transistor Q_i31 transfers the signal stored in the previous bit-level cell M_in, further delayed by the delay time t_d2 determined by the second delay element D_i32 to the capacitor C_i3. When the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrodes of the first nMOS transistor Q_i31and the drain electrodes of the second nMOS transistor Q_i32 cannot deliver the signals, which are entered to the gate electrodes of the first nMOS transistor Q_i31, further to the next bit-level cell M_i2, at a time when time proceeds 1/2TAU_clock, as the signals is blocked to be transferred to the gate electrodes of the next first nMOS transistor Q_i21 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D_i22

And, as illustrated in Fig. 25(a), in a reverse-directional marching main memory, a bit-level cell M_i2 of the second column from the left, and on the i-th row, encompasses a first nMOS transistor Q_i21 having a drain electrode connected to the clock signal supply line through a first delay element D_i21 and a gate electrode connected to the output terminal of the bit-level cell M_i3 through a second delay element D_i22; a second nMOS transistor Q_i22 having a drain electrode connected to a source electrode of the first nMOS transistor Q_i21, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i2configured to store the information of the bit-level cell M_i2, connected in parallel with the second nMOS transistor Q_i22. When the clock signal becomes the logical level of "1", the second nMOS transistor Q_i22 begins to discharge the signal charge, which is already stored in the capacitor C_i2 at a previous clock cycle. After the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_i2 is completely discharged to becomes the logical level of "0", the first nMOS transistor Q_i21 becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first delay element D_i21. Thereafter, when the signal is fed from the output terminal of the bit-level cell M_i3 to the gate electrode of the first nMOS transistor Q_i21, the first nMOS transistor Q_i21 transfers the signal stored in the previous bit-level cell M_i3, further delayed by the delay time t_d2 determined by the second delay element D_i22 to the capacitor C_i2. When the clock signal becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first nMOS transistor Q_i21and the drain electrode of the second nMOS transistor Q_i22 cannot deliver the signal, which is entered to the gate electrode of the first nMOS transistor Q_i21, further to the next bit-level cell M_i1, at a time when time proceeds 1/2TAU_clock, as the signal is blocked to be transferred to the gate electrode of the next first nMOS transistor Q_i11 delayed by the delay time t_d2 = 1/2TAU_clock determined by the second delay element D_i12.

As illustrated in Fig. 25(a), in a reverse-directional marching main memory, a bit-level cell M_i1 of the first column and on the i-th row, , which is allocated at the leftmost side on the i-th row and connected to an output terminal OUT, encompasses a first nMOS transistor Q_i11 having a drain electrode connected to the clock signal supply line through a first delay element D_i11 and a gate electrode connected to the output terminal of the bit-level cell M_i2 through a second delay element D_i12; a second nMOS transistor Q_i12 having a drain electrode connected to a source electrode of the first nMOS transistor Q_i11, a gate electrode connected to the clock signal supply line, and a source electrode connected to the ground potential; and a capacitor C_i1configured to store the information of the bit-level cell M_i1, connected in parallel with the second nMOS transistor Q_i12. When the clock signal becomes the logical level of "1", the second nMOS transistor Q_i12 begins to discharge the signal charge, which is already stored in the capacitor C_i1 at a previous clock cycle. After the clock signal of the logical level of "1" is applied and the signal charge stored in the capacitor C_i1 is completely discharged to becomes the logical level of "0", the first nMOS transistor Q_i11 becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first delay element D_i11. Thereafter, when the signal is fed from the output terminal of the bit-level cell M_i2 to the gate electrode of the first nMOS transistor Q_i11, the first nMOS transistor Q_i11 transfers the signal stored in the previous bit-level cell M_i2, further delayed by the delay time t_d2 determined by the second delay element D_i12 to the capacitor C_i1. The output node connecting the source electrode of the first nMOS transistor Q_i11and the drain electrode of the second nMOS transistor Q_i12 delivers the signal stored in the capacitor C_i1 to the output terminal OUT.

According to the reverse-directional one-dimensional marching main memory 31 of the first embodiment illustrated in Figs. 24. 25(a) and 25(b), addressing to each of memory units U₁, U₂, U₃,........., U_n-1, U_n disappears and required information is heading for its destination unit connected to the edge of the memory. The mechanism of accessing the reverse-directional one-dimensional marching main memory 31 of the first embodiment is truly alternative to existing memory schemes that are starting from the addressing mode to read/write information. Therefore, according to the reverse-directional one-dimensional marching main memory 31 of the first embodiment, the memory-accessing without addressing mode is quite simpler than existing memory schemes.

As above mentioned, the bit-level cell M_ij can establish "a marching AND-gate" operation. Therefore, as illustrated in Fig. 26, in a gate-level representation of the cell array corresponding to the reverse-directional marching main memory 31 illustrated in Fig. 25(a), the n-th bit-level cell M_i,n allocated at the rightmost side on the i-th row and connected to an input terminal IN encompasses a capacitor C_inconfigured to store the information, and a marching AND-gate G_inhaving one input terminal connected to the capacitor C_in, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the preceding marching AND-gate G_in-1 assigned to the adjacent (n-1)-th bit-level cell M_i,n-1 on the i-th row. When the logical values of "1" is fed to the other input terminal of the marching AND-gate G_n, the information stored in the capacitor C_in is transferred to a capacitor C_i,n-1, assigned to the adjacent (n-1)-th bit-level cell M_i,n-1 on the i-th row, and the capacitor C_i,n-1stores the information. Namely, the (n-1)-th bit-level cell M_i,n-1 on the i-th row of the reverse-directional marching main memory encompasses the capacitor C_i,n-1and a marching AND-gate G_i,n-1, which has one input terminal connected to the capacitor C_i,n-1, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the preceding marching AND-gate G_i,n-2 assigned to the adjacent third bit-level cell M_i,n-2(illustration is omitted).

Similarly the third bit-level cell M_i3 on the i-th row of the reverse-directional marching main memory encompasses a capacitor C_i3configured to store the information, and a marching AND-gate G_i3having one input terminal connected to the capacitor C_i3, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the preceding marching AND-gate G_i2 assigned to the adjacent second bit-level cell M_i2. Therefore, when the logical values of "1" is fed to the other input terminal of the marching AND-gate G_i3, the information stored in the capacitor C_i3 is transferred to the capacitor C_i2, assigned to the second bit-level cell M_i2, and the capacitor C_i2stores the information.

Furthermore, the second bit-level cell M_i2 on the i-th row of the reverse-directional marching main memory encompasses the capacitor C_i2configured to store the information, and the marching AND-gate G_i2having one input terminal connected to the capacitor C_i2, the other input terminal configured to be supplied with the clock signal, and an output terminal connected to one input terminal of the preceding marching AND-gate G_i1 assigned to the adjacent first bit-level cell M_i1, which is allocated at the leftmost side on the i-th row and connected to an output terminal OUT.

The concept of marching main memory 31 used in the computer system pertaining to the first embodiment of the present invention is illustrated in Fig. 27, this is different from existing computer memory, because the marching main memory 31 is purposely designed with functionality of storage and conveyance of information/data through all of memory units U₁, U₂, U₃,........., U_n-1, U_n in the marching main memory 31. Marching memory supplies information/data to the processor (CPU) 11 at the same speed of the processor 11. As illustrated in the time-domain relationship of Fig.9, the memory unit streaming time T_mus required for transferring information/data through one memory units U₁, U₂, U₃,........., U_n-1, U_n, in the marching main memory 31 is equal to the clock cycle T_cc in the processor 11. The marching main memory 31 stores information/data in each of the memory units U₁, U₂, U₃,........., U_n-1, U_n, and transfers synchronously with the clock signal, step by step, toward the output terminals, so as to provide the processor 11 with the stored information/data so that the arithmetic logic unit 112 can execute the arithmetic and logic operations with the stored information/data.

Therefore, as illustrated in Fig. 28, marching memory structure 3 includes the marching main memory 31 of the first embodiment of the present invention. The term of "the marching memory structure 3" means a generic concept of the memory structure including a marching-instruction register file (RF) 22a and a marching-data register file (RF) 22b connected to the ALU 112, which will be explained further in the following second embodiment, and a marching-instruction cache memory 21a and a marching-data cache memory 21b, which will be explained further in the following third embodiment, in addition to the marching main memory 31 used in the computer system pertaining to the first embodiment of the present invention.

Fig. 29(a) illustrates a forward data-stream S_f flowing from the marching memory structure 3 to the processor 11 and backward data-stream (reverse data-stream) S_b flowing from the processor 11 to the marching memory structure 3, and Fig. 29(a) illustrates bandwidths established between the marching memory structure 3 and the processor 11 assuming that the memory unit streaming time T_mus in the marching memory structure 3 is equal to the clock cycle T_cc of the processor 11.

The scheme of the marching main memory 31 may be considered to be analogous to a magnetic tape system illustrated in Fig. 30 (a), which encompasses a magnetic tape 503, a take-up reel 502 for winding the magnetic tape 503, a supply reel 501 for rewinding and releasing the magnetic tape 503, a read/write header 504 for reading information/data from the magnetic tape 503 or writing information/data to the magnetic tape 503, and a processor 11 connected to the read/write header 504. As the take-up reel 502 winds the magnetic tape 503, which is released from the supply reel 501, the magnetic tape 503 moves at high speed from the supply reel 501 toward the take-up reel 502, and information/data stored on the magnetic tape 503, being transferred with the movement of the magnetic tape 503 at high speed, are read by the read/write header 504. And the processor 11 connected to the read/write header 504 can execute arithmetic and logic operations with information/data read from the magnetic tape 503. Alternatively, the results of the processing in the processor 11 are sent out to the magnetic tape 503 through the read/write header 504.

If we suppose the architecture of the magnetic tape system illustrated in Fig. 30 (a) is implemented by semiconductor technology, that is, If we image an extremely high-speed magnetic tape system is virtually established on semiconductor silicon chip as illustrated in Fig. 30 (b), the extremely high-speed magnetic tape system illustrated in Fig. 30 (a) may correspond to a net marching memory structure 3, including the marching main memory 31 of the first embodiment of the present invention. The net marching memory structure 3 illustrated in Fig. 30 (b) stores information/data in each of the memory units on the silicon chip and transfers synchronously with the clock signal, step by step, toward the take-up reel 502, so as to provide the processor 11 with the stored information/data actively and sequentially so that the processor 11 can execute the arithmetic and logic operations with the stored information/data, and the results of the processing in the processor 11 are sent out to the net marching memory structure 3.

(BIDIRECTIONAL MARCHING MAIN MEMORY)
As illustrated in Figs. 31 (a)-(c), the marching main memory 31 of the first embodiment of the present invention, can achieve bidirectional transferring of information/data. That is, Fig. 31 (a) illustrates a forward marching behavior of information/data, in which information/data marches (shifts) side by side toward right-hand direction (forward direction) in a one-dimensional marching main memory 31, Fig. 31 (b) illustrates a staying state of the one-dimensional marching main memory 31, and Fig. 31 (c) illustrates a reverse-marching behavior of information/data (a backward marching behavior), in which information/data marches (shifts) side by side toward left-hand direction (reverse direction) in the one-dimensional marching main memory 31.

Figs. 32 and 33 illustrate two examples of the representative arrays of i-th row of the m * n matrix (here, "m" is an integer determined by word size) in a transistor-level representation of the cell array for the bidirectional marching main memory 31, respectively, which can achieve the bidirectional behavior illustrated in Figs. 31 (a)-(c). The bidirectional marching main memory 31 stores the information/data of bit level in each of cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,n and transfers bi-directionally the information/data synchronously with the clock signal, step by step in the forward direction and/or reverse direction (backward direction) between a first I/O selector 512 and a second I/O selector 513.

In Figs. 32 and 33, each of the cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,nis assigned in memory unit U₁, U₂, U₃,........., U_n-1, U_n, respectively. That is the cell M_i1is assigned as the first bit-level cell in the first memory unit U₁, the first memory unit U₁ stores information of byte size or word size by the sequence of bit-level cells arrayed in the first memory unit U₁. Similarly, the cell M_i2is assigned as the second bit-level cell in the second memory unit U₂, the cell M_i3is assigned as the third bit-level cell in the third memory unit U₃, ........., the cell M_i,n-1is assigned as the (n-1)-th bit-level cell in the (n-1)-th memory unit U_n-1, and the cell M_i,nis assigned as the n-th bit-level cell in the n-th memory unit U_n. And the memory units U₂, U₃,........., U_n-1, U_n stores information of byte size or word size by the sequence of bit-level cells arrayed in the memory unit U₂, U₃,........., U_n-1, U_n, respectively. Therefore, the bidirectional marching main memory 31 stores the information/data of byte size or word size in each of cells U₁, U₂, U₃,........., U_n-1, U_n and transfers bi-directionally the information/data of byte size or word size synchronously with the clock signal, pari passu, in the forward direction and/or reverse direction (backward direction) between a first I/O selector 512 and a second I/O selector 513.

A clock selector 511 selects a first clock signal supply line CL1 and a second clock signal supply line CL2. The first clock signal supply line CL1 drives the forward data-stream, and the second clock signal supply line CL2 drives the backward data-stream, and each of the first clock signal supply line CL1 and the second clock signal supply line CL2 has logical values of "1" and "0".

In the transistor-level representation of the cell array implementing the marching main memory 31 illustrated in Fig. 32, a first bit-level cell M_i1 allocated at the leftmost side on i-th row, being connected to a first I/O selector 512, encompasses a first forward nMOS transistor Q_i11f having a drain electrode connected to a first clock signal supply line CL1 through a first forward delay element D_i11f and a gate electrode connected to the first I/O selector 512 through a second forward delay element D_i12f; a second forward nMOS transistor Q_i12f having a drain electrode connected to a source electrode of the first forward nMOS transistor Q_i11f, a gate electrode connected to the first clock signal supply line, and a source electrode connected to the ground potential; and a forward capacitor C_i1fconfigured to store the forward information/data of the cell M_i1, connected in parallel with the second forward nMOS transistor Q_i12f, wherein an output node connecting the source electrode of the first forward nMOS transistor Q_i11f and the drain electrode of the second forward nMOS transistor Q_i12f serves as a forward output terminal of the cell M_i1, configured to transfer the signal stored in the forward capacitor C_i1f to the next bit-level cell M_i2. The first bit-level cell M_i1 further encompasses a first backward nMOS transistor Q_i11g having a drain electrode connected to a second clock signal supply line through a first backward delay element D_i11g and a gate electrode connected to the backward output terminal of the bit-level cell M_i2 through a second backward delay element D_i12g; a second backward nMOS transistor Q_i12g having a drain electrode connected to a source electrode of the first backward nMOS transistor Q_i11g, a gate electrode connected to the second clock signal supply line, and a source electrode connected to the ground potential; and a backward capacitor C_i1gconfigured to store the backward information/data of the cell M_i1, connected in parallel with the second backward nMOS transistor Q_i12g, wherein an output node connecting the source electrode of the first backward nMOS transistor Q_i11g and the drain electrode of the second backward nMOS transistor Q_i12g serves as a backward output terminal of the cell M_i1, configured to transfer the signal stored in the backward capacitor C_i1g to the first I/O selector 512.

A second bit-level cell M_i2 allocated at the second from the left side on i-th row, being connected to the bit-level cell M_i1, encompasses a first forward nMOS transistor Q_i21f having a drain electrode connected to the first clock signal supply line CL1 through a first forward delay element D_i21f and a gate electrode connected to the forward output terminal of the bit-level cell M_i1through a second forward delay element D_i22f; a second forward nMOS transistor Q_i22f having a drain electrode connected to a source electrode of the first forward nMOS transistor Q_i21f, a gate electrode connected to the first clock signal supply line CL1, and a source electrode connected to the ground potential; and a forward capacitor C_i2fconfigured to store the forward information/data of the cell M_i2, connected in parallel with the second forward nMOS transistor Q_i22f ,wherein an output node connecting the source electrode of the first forward nMOS transistor Q_i21f and the drain electrode of the second forward nMOS transistor Q_i22f serves as a forward output terminal of the cell M_i2, configured to transfer the signal stored in the forward capacitor C_i2f to the next bit-level cell M_i3. The second bit-level cell M_i2 further encompasses a first backward nMOS transistor Q_i21g having a drain electrode connected to the second clock signal supply line CL2 through a first backward delay element D_i21g and a gate electrode connected to the backward output terminal of the bit-level cell M_i3 through a second backward delay element D_i22g; a second backward nMOS transistor Q_i22g having a drain electrode connected to a source electrode of the first backward nMOS transistor Q_i21g, a gate electrode connected to the second clock signal supply line CL2, and a source electrode connected to the ground potential; and a backward capacitor C_i2gconfigured to store the backward information/data of the cell M_i2, connected in parallel with the second backward nMOS transistor Q_i22g ,wherein an output node connecting the source electrode of the first backward nMOS transistor Q_i21g and the drain electrode of the second backward nMOS transistor Q_i22g serves as a backward output terminal of the cell M_i2, configured to transfer the signal stored in the backward capacitor C_i2g to the next bit-level cell M_i1.

A third bit-level cell M_i3 allocated at the second from the left side on i-th row, being connected to the bit-level cell M_i2, encompasses a first forward nMOS transistor Q_i31f having a drain electrode connected to the first clock signal supply line CL1 through a first forward delay element D_i31f and a gate electrode connected to the forward output terminal of the bit-level cell M_i2through a second forward delay element D_i32f; a second forward nMOS transistor Q_i32f having a drain electrode connected to a source electrode of the first forward nMOS transistor Q_i31f, a gate electrode connected to the first clock signal supply line CL1, and a source electrode connected to the ground potential; and a forward capacitor C_i3fconfigured to store the forward information/data of the cell M_i3, connected in parallel with the second forward nMOS transistor Q_i32f, wherein an output node connecting the source electrode of the first forward nMOS transistor Q_i31f and the drain electrode of the second forward nMOS transistor Q_i32f serves as a forward output terminal of the cell M_i3, configured to transfer the signal stored in the forward capacitor C_i3f to the next bit-level cell M_i4(illustration is omitted). The third bit-level cell M_i3 further encompasses a first backward nMOS transistor Q_i31g having a drain electrode connected to the second clock signal supply line CL2 through a first backward delay element D_i31g and a gate electrode connected to the backward output terminal of the bit-level cell M_i4 through a second backward delay element D_i32g; a second backward nMOS transistor Q_i32g having a drain electrode connected to a source electrode of the first backward nMOS transistor Q_i31g, a gate electrode connected to the second clock signal supply line CL2, and a source electrode connected to the ground potential; and a backward capacitor C_i3gconfigured to store the backward information/data of the cell M_i3, connected in parallel with the second backward nMOS transistor Q_i32g ,wherein an output node connecting the source electrode of the first backward nMOS transistor Q_i31g and the drain electrode of the second backward nMOS transistor Q_i32g serves as a backward output terminal of the cell M_i3, configured to transfer the signal stored in the backward capacitor C_i3g to the next bit-level cell M_i2.

A (n-1)-th bit-level cell M_i(n-1) allocated at the second from the left side on i-th row, encompasses a first forward nMOS transistor Q_i(n-1)1f having a drain electrode connected to the first clock signal supply line CL1 through a first forward delay element D_i(n-1)1f and a gate electrode connected to the forward output terminal of the bit-level cell M_i(n-2)(illustration is omitted) through a second forward delay element D_i(n-1)2f; a second forward nMOS transistor Q_i(n-1)2f having a drain electrode connected to a source electrode of the first forward nMOS transistor Q_i(n-1)1f, a gate electrode connected to the first clock signal supply line CL1, and a source electrode connected to the ground potential; and a forward capacitor C_i(n-1)fconfigured to store the forward information/data of the cell M_i(n-1), connected in parallel with the second forward nMOS transistor Q_i(n-1)2f, wherein an output node connecting the source electrode of the first forward nMOS transistor Q_i(n-1)1f and the drain electrode of the second forward nMOS transistor Q_i(n-1)2f serves as a forward output terminal of the cell M_i(n-1), configured to transfer the signal stored in the forward capacitor C_i(n-1)f to the next bit-level cell M_in. The (n-1)-th bit-level cell M_i(n-1) further encompasses a first backward nMOS transistor Q_i(n-1)1g having a drain electrode connected to the second clock signal supply line CL2 through a first backward delay element D_i(n-1)1g and a gate electrode connected to the backward output terminal of next bit-level cell M_in through a second backward delay element D_i(n-1)2g; a second backward nMOS transistor Q_i(n-1)2g having a drain electrode connected to a source electrode of the first backward nMOS transistor Q_i(n-1)1g, a gate electrode connected to the second clock signal supply line CL2, and a source electrode connected to the ground potential; and a backward capacitor C_i(n-1)gconfigured to store the backward information/data of the cell M_i(n-1), connected in parallel with the second backward nMOS transistor Q_i(n-1)2g ,wherein an output node connecting the source electrode of the first backward nMOS transistor Q_i(n-1)1g and the drain electrode of the second backward nMOS transistor Q_i(n-1)2g serves as a backward output terminal of the cell M_i(n-1), configured to transfer the signal stored in the backward capacitor C_i(n-1)g to the next bit-level cell M_i(n-2) (illustration is omitted).

A n-th bit-level cell M_in allocated at the rightmost side on i-th row, encompasses a first forward nMOS transistor Q_in1f having a drain electrode connected to the first clock signal supply line CL1 through a first forward delay element D_in1f and a gate electrode connected to the forward output terminal of the bit-level cell M_i(n-1)through a second forward delay element D_in2f; a second forward nMOS transistor Q_in2f having a drain electrode connected to a source electrode of the first forward nMOS transistor Q_in1f, a gate electrode connected to the first clock signal supply line CL1, and a source electrode connected to the ground potential; and a forward capacitor C_infconfigured to store the forward information/data of the cell M_in, connected in parallel with the second forward nMOS transistor Q_in2f, wherein an output node connecting the source electrode of the first forward nMOS transistor Q_in1f and the drain electrode of the second forward nMOS transistor Q_in2f serves as a forward output terminal of the cell M_in, configured to transfer the signal stored in the forward capacitor C_inf to the second I/O selector 513. The n-th bit-level cell M_in further encompasses a first backward nMOS transistor Q_in1g having a drain electrode connected to the second clock signal supply line CL2 through a first backward delay element D_in1g and a gate electrode connected to the second I/O selector 513 through a second backward delay element D_in2g; a second backward nMOS transistor Q_in2g having a drain electrode connected to a source electrode of the first backward nMOS transistor Q_in1g, a gate electrode connected to the second clock signal supply line CL2, and a source electrode connected to the ground potential; and a backward capacitor C_ingconfigured to store the backward information/data of the cell M_in, connected in parallel with the second backward nMOS transistor Q_in2g ,wherein an output node connecting the source electrode of the first backward nMOS transistor Q_in1g and the drain electrode of the second backward nMOS transistor Q_in2g serves as a backward output terminal of the cell M_in, configured to transfer the signal stored in the backward capacitor C_ing to the next bit-level cell M_i(n-1).

When the clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "1", the second forward nMOS transistor Q_i12f in the first memory unit U₁ begin to discharge the signal charge, which is already stored in the forward capacitor C_i1f in the first memory unit U₁ at a previous clock cycle. And, after the clock signal of the logical level of "1", supplied from the first clock signal supply line CL1, is applied to the second forward nMOS transistor Q_i12f,and the signal charge stored in the forward capacitor C_i1f is completely discharged to becomes the logical level of "0", the first forward nMOS transistor Q_i11f, becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first forward delay element D_i11f. Thereafter, when the information/data of bit level is entered from the first I/O selector 512 to the gate electrode of the first forward nMOS transistor Q_i11f, the first forward nMOS transistor Q_i11ftransfers the information/data to the forward capacitor C_i1f, delayed by the delay time t_d2 determined by the second forward delay element D_i12f. When the clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first forward nMOS transistor Q_i11fand the drain electrode of the second forward nMOS transistor Q_i12f cannot deliver the information/data, which is entered from the first I/O selector 512 to the gate electrode of the first forward nMOS transistor Q_i11f, further to the next bit-level cell M_i2, at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first forward nMOS transistor Q_i21f delayed by the delay time t_d2 = 1/2TAU_clock determined by the second forward delay element D_i22f.

When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "1", the second backward nMOS transistor Q_i12b begins to discharge the signal charge, which is already stored in the backward capacitor C_i1b at a previous clock cycle. After the clock signal of the logical level of "1", supplied from the second clock signal supply line CL2, is applied and the signal charge stored in the backward capacitor C_i1b is completely discharged to becomes the logical level of "0", the first backward nMOS transistor Q_i11b becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first backward delay element D_i11b. Thereafter, when the information/data is fed from the backward output terminal of the bit-level cell M_i2 to the gate electrode of the first backward nMOS transistor Q_i11b, the first backward nMOS transistor Q_i11b transfers the information/data stored in the previous bit-level cell M_i2, further delayed by the delay time t_d2 determined by the second backward delay element D_i12b to the backward capacitor C_i1b. The output node connecting the source electrode of the first backward nMOS transistor Q_i11band the drain electrode of the second backward nMOS transistor Q_i12b delivers the information/data stored in the backward capacitor C_i1b to the first I/O selector 512.

And, when the next clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "1", and the second forward nMOS transistor Q_i22f in the second memory unit U₂ begin to discharge the signal charge, which is already stored in the forward capacitor C_i2f in the second memory unit U₂ at the previous clock cycle. And, after the clock signal of the logical level of "1", supplied from the first clock signal supply line CL1, is applied to the second forward nMOS transistor Q_i22f,and the signal charge stored in the forward capacitor C_i2f is completely discharged to becomes the logical level of "0", the first forward nMOS transistor Q_i2f1 becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first forward delay element D_i21f. Thereafter, when the information/data of bit level stored in the previous forward capacitor C_i1f is fed to the gate electrode of the first forward nMOS transistor Q_i21f and the first forward nMOS transistor Q_i21f transfers the information/data, delayed by the delay time t_d2 determined by the second forward delay element D_i22f to the forward capacitor C_i2f. When the clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first forward nMOS transistor Q_i21fand the drain electrode of the second forward nMOS transistor Q_i22f cannot deliver the information/data, which is entered to the gate electrode of the first forward nMOS transistor Q_i21f, further to the next bit-level cell M_i3, at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first forward nMOS transistor Q_i31f delayed by the delay time t_d2 = 1/2TAU_clock determined by the second forward delay element D_i32f.

When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "1", the second backward nMOS transistor Q_i22b begins to discharge the signal charge, which is already stored in the backward capacitor C_i2b at a previous clock cycle. After the clock signal supplied from the second clock signal supply line CL2 of the logical level of "1" is applied and the signal charge stored in the backward capacitor C_i2b is completely discharged to becomes the logical level of "0", the first backward nMOS transistor Q_i21b becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first backward delay element D_i21b. Thereafter, when the information/data is fed from the backward output terminal of the bit-level cell M_i3 to the gate electrode of the first backward nMOS transistor Q_i21b, the first backward nMOS transistor Q_i21b transfers the information/data stored in the previous bit-level cell M_i3, further delayed by the delay time t_d2 determined by the second backward delay element D_i22b to the backward capacitor C_i2b. When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first backward nMOS transistor Q_i21band the drain electrode of the second backward nMOS transistor Q_i22b cannot deliver the information/data, which is entered to the gate electrode of the first backward nMOS transistor Q_i21b, further to the next bit-level cell M_i1, at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first backward nMOS transistor Q_i11b delayed by the delay time t_d2 = 1/2TAU_clock determined by the second backward delay element D_i12b.

And, when the next clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "1", the second forward nMOS transistor Q_i32f in the third memory unit U₃ begin to discharge the signal charge, which is already stored in the forward capacitor C_i3f in the third memory unit U₃ at the previous clock cycle. And, after the clock signal of the logical level of "1", supplied from the first clock signal supply line CL1, is applied to the second forward nMOS transistor Q_i32f,and the signal charge stored in the forward capacitor C_i3f is completely discharged to becomes the logical level of "0", the first forward nMOS transistor Q_i31f becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first forward delay element D_i31f. Thereafter, when the information/data stored in the previous forward capacitor C_i2f is fed to the gate electrode of the first forward nMOS transistor Q_i31f, and the first forward nMOS transistor Q_i31f transfers the information/data, delayed by the delay time t_d2 determined by the second forward delay element D_i32f to the forward capacitor C_i3f. When the clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first forward nMOS transistor Q_i31fand the drain electrode of the second forward nMOS transistor Q_i32f cannot deliver the information/data, which is entered to the gate electrode of the first forward nMOS transistor Q_i31f, further to the next bit-level cell M_i4(illustration is omitted), at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first forward nMOS transistor Q_i41f (illustration is omitted) delayed by the delay time t_d2 = 1/2TAU_clock determined by the second forward delay element D_i42f(illustration is omitted).

When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "1", the second backward nMOS transistor Q_i32b begins to discharge the signal charge, which is already stored in the backward capacitor C_i3b at a previous clock cycle. After the clock signal supplied from the second clock signal supply line CL2 of the logical level of "1" is applied and the signal charge stored in the backward capacitor C_i3b is completely discharged to becomes the logical level of "0", the first backward nMOS transistor Q_i31b becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first backward delay element D_i31b. Thereafter, when the information/data is fed from the backward output terminal of the bit-level cell M_i3 to the gate electrode of the first backward nMOS transistor Q_i31b, the first backward nMOS transistor Q_i31b transfers the information/data stored in the previous bit-level cell M_i3, further delayed by the delay time t_d2 determined by the second backward delay element D_i32b to the backward capacitor C_i3b. When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first backward nMOS transistor Q_i31band the drain electrode of the second backward nMOS transistor Q_i32b cannot deliver the information/data, which is entered to the gate electrode of the first backward nMOS transistor Q_i31b, further to the next bit-level cell M_i2, at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first backward nMOS transistor Q_i21b delayed by the delay time t_d2 = 1/2TAU_clock determined by the second backward delay element D_i22b.

And, when the next clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "1", the second forward nMOS transistor Q_i(n-1)2f in the third memory unit U_(n-1) begin to discharge the signal charge, which is already stored in the forward capacitor C_i(n-1)f in the third memory unit U_(n-1) at the previous clock cycle. And, after the clock signal of the logical level of "1", supplied from the first clock signal supply line CL1, is applied to the second forward nMOS transistor Q_i(n-1)2f,and the signal charge stored in the forward capacitor C_i(n-1)f is completely discharged to becomes the logical level of "0", the first forward nMOS transistor Q_i(n-1)1f becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first forward delay element D_i(n-1)1f. Thereafter, when the information/data stored in the previous forward capacitor C_i2f is fed to the gate electrode of the first forward nMOS transistor Q_i(n-1)1f, and the first forward nMOS transistor Q_i(n-1)1f transfers the information/data, delayed by the delay time t_d2 determined by the second forward delay element D_i(n-1)2f to the forward capacitor C_i(n-1)f. When the clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first forward nMOS transistor Q_i(n-1)1fand the drain electrode of the second forward nMOS transistor Q_i(n-1)2f cannot deliver the information/data, which is entered to the gate electrode of the first forward nMOS transistor Q_i(n-1)1f, further to the next bit-level cell M_in, at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first forward nMOS transistor Q_in1f delayed by the delay time t_d2 = 1/2TAU_clock determined by the second forward delay element D_in2f.

When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "1", the second backward nMOS transistor Q_i(n-1)2b begins to discharge the signal charge, which is already stored in the backward capacitor C_i(n-1)b at a previous clock cycle. After the clock signal supplied from the second clock signal supply line CL2 of the logical level of "1" is applied and the signal charge stored in the backward capacitor C_i(n-1)b is completely discharged to becomes the logical level of "0", the first backward nMOS transistor Q_i(n-1)1b becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first backward delay element D_i(n-1)1b. Thereafter, when the information/data is fed from the backward output terminal of the bit-level cell M_i(n-1) to the gate electrode of the first backward nMOS transistor Q_i(n-1)1b, the first backward nMOS transistor Q_i(n-1)1b transfers the information/data stored in the previous bit-level cell M_i(n-1), further delayed by the delay time t_d2 determined by the second backward delay element D_i(n-1)2b to the backward capacitor C_i(n-1)b. When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first backward nMOS transistor Q_i(n-1)1band the drain electrode of the second backward nMOS transistor Q_i(n-1)2b cannot deliver the information/data, which is entered to the gate electrode of the first backward nMOS transistor Q_i(n-1)1b, further to the next bit-level cell M_i(n-2) (illustration is omitted), at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first backward nMOS transistor Q_i(n-2)1b (illustration is omitted)delayed by the delay time t_d2 = 1/2TAU_clock determined by the second backward delay element D_i(n-2)2b(illustration is omitted).

And, when the next clock signal supplied from the first clock signal supply line CL1 becomes the logical level of "1", the second forward nMOS transistor Q_in2f in the third memory unit U_n begin to discharge the signal charge, which is already stored in the forward capacitor C_inf in the third memory unit U_n at the previous clock cycle. And, after the clock signal of the logical level of "1", supplied from the first clock signal supply line CL1, is applied to the second forward nMOS transistor Q_in2f,and the signal charge stored in the forward capacitor C_inf is completely discharged to becomes the logical level of "0", the first forward nMOS transistor Q_in1f becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first forward delay element D_in1f. Thereafter, when the information/data stored in the previous forward capacitor C_i2f is fed to the gate electrode of the first forward nMOS transistor Q_in1f, and the first forward nMOS transistor Q_in1f transfers the information/data, delayed by the delay time t_d2 determined by the second forward delay element D_in2f to the forward capacitor C_inf. The output node connecting the source electrode of the first forward nMOS transistor Q_in1fand the drain electrode of the second forward nMOS transistor Q_in2f delivers the information/data, which is entered to the gate electrode of the first forward nMOS transistor Q_in1f to the second I/O selector 513.

When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "1", the second backward nMOS transistor Q_in2b begins to discharge the signal charge, which is already stored in the backward capacitor C_inb at a previous clock cycle. After the clock signal supplied from the second clock signal supply line CL2 of the logical level of "1" is applied and the signal charge stored in the backward capacitor C_inb is completely discharged to becomes the logical level of "0", the first backward nMOS transistor Q_in1b becomes active as the transfer transistor, delayed by the delay time t_d1 determined by the first backward delay element D_in1b. Thereafter, when the information/data is fed from the second I/O selector 513 to the gate electrode of the first backward nMOS transistor Q_in1b, the first backward nMOS transistor Q_in1b transfers the information/data received from the second I/O selector 513, further delayed by the delay time t_d2 determined by the second backward delay element D_in2b to the backward capacitor C_inb. When the clock signal supplied from the second clock signal supply line CL2 becomes the logical level of "0" at a time when time proceeds 1/2TAU_clock, the output node connecting the source electrode of the first backward nMOS transistor Q_in1band the drain electrode of the second backward nMOS transistor Q_in2b cannot deliver the information/data, which is entered to the gate electrode of the first backward nMOS transistor Q_in1b, further to the next bit-level cell M_i(n-2) (illustration is omitted), at a time when time proceeds 1/2TAU_clock, as the information/data is blocked to be transferred to the gate electrode of the next first backward nMOS transistor Q_i(n-2)1b (illustration is omitted)delayed by the delay time t_d2 = 1/2TAU_clock determined by the second backward delay element D_i(n-2)2b(illustration is omitted).

Therefore, in the bidirectional marching main memory illustrated in Fig. 32, each of the cells M_i1, M_i2, M_i3,........., M_i,(n-1), M_i,n on the i-th row of the bidirectional marching main memory stores the information/data, and transfers bi-directionally the information/data, synchronously with the clock signals supplied respectively from the first clock signal supply line CL1 and the second clock signal supply line CL2, step by step, between the first I/O selector 512 and the second I/O selector 513. As already explained, because each of the cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,nis assigned in memory unit U₁, U₂, U₃,........., U_n-1, U_n, respectively, and the memory units U₂, U₃,........., U_n-1, U_n stores information of byte size or word size by the sequence of bit-level cells arrayed in the memory unit U₂, U₃,........., U_n-1, U_n, respectively, the bidirectional marching main memory 31 illustrated in Fig.32 stores the information/data of byte size or word size in each of cells U₁, U₂, U₃,........., U_n-1, U_n and transfers bi-directionally the information/data of byte size or word size synchronously with the clock signal, pari passu, in the forward direction and/or reverse direction (backward direction) between a first I/O selector 512 and a second I/O selector 513, so as to provide the processor 11 with the stored information/data of byte size or word size actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information/data.

As illustrated in Fig. 33, a forward isolation transistor Q_i23f is provided so as to isolate the signal-storage state of the second bit-level cell M_i2 in the second memory unit U_n from the signal-storage state of the first bit-level cell M_i1 in the first memory unit U₁, the forward isolation transistor Q_i23f transfers forward a signal from the first bit-level cell M_i1 to the second bit-level cell M_i2 at a required timing determined by a clock signal, which is supplied through the first clock signal supply line CL1. And, a backward isolation transistor Q_i13b is provided so as to isolate the signal-storage state of the signal-storage state of the first bit-level cell M_i1 in the first memory unit U₁ from the second bit-level cell M_i2 in the second memory unit U₂, the backward isolation transistor Q_i13b transfers backward a signal from the second bit-level cell M_i2 to the first bit-level cell M_i1 at a required timing determined by a clock signal, which is supplied through the second clock signal supply line CL2. Then, a sequence of the forward isolation transistors Q_i23f (i=1 to m; "m" is integer corresponding the byte size or the word size) arrayed in parallel with the memory units U₁ and U₂ transfers forward the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL1 so that the information of byte size or word size can march along the forward direction, pari passu. And, a sequence of the backward isolation transistors Q_i13b (i=1 to m) arrayed in parallel with the memory units U₁ and U₂ transfers backward the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL2 so that the information of byte size or word size can march along the backward direction, pari passu.

And, similarly, a backward isolation transistor Q_i23b is provided so as to isolate the signal-storage state of the signal-storage state of the second bit-level cell M_i2 in the second memory unit U₂ from the third bit-level cell M_i3 (the illustration is omitted) in the third memory unit U₃, the backward isolation transistor Q_i23b transfers backward a signal from the third bit-level cell M_i3 to the second bit-level cell M_i2 at a required timing determined by a clock signal, which is supplied through the third clock signal supply line CL2. And, a sequence of the backward isolation transistors Q_i23b (i=1 to m) arrayed in parallel with the memory units U₂ and U₃ transfers backward the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL2 so that the information of byte size or word size can march along the backward direction, pari passu.

Furthermore, as illustrated in Fig. 33, a forward isolation transistor Q_i(n-1)3f is provided so as to isolate the signal-storage state of the (n-1)-th bit-level cell M_i(n-1) in the (n-1)-th memory unit U_n-1 from the signal-storage state of the (n-2)-th bit-level cell M_i(n-2) (the illustration is omitted) in the (n-2)-th memory unit U_n-2(the illustration is omitted), the forward isolation transistor Q_i(n-1)3f transfers forward a signal from the (n-2)-th bit-level cell M_i(n-2) to the (n-1)-th bit-level cell M_i(n-1) at a required timing determined by a clock signal, which is supplied through the first clock signal supply line CL1. Then, a sequence of the forward isolation transistors Q_i(n-1)3f (i=1 to m) arrayed in parallel with the memory units U_n-2 and U_n-1 transfers the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL1 so that the information of byte size or word size can march along the forward direction, pari passu.

And, a forward isolation transistor Q_in3f is provided so as to isolate the signal-storage state of the n-th bit-level cell M_in in the n-th memory unit U_n from the signal-storage state of the (n-1)-th bit-level cell M_in-1 in the (n-1)-th memory unit U_n-1, the forward isolation transistor Q_in3f transfers forward a signal from the (n-1)-th bit-level cell M_in-1 to the n-th bit-level cell M_in at a required timing determined by a clock signal, which is supplied through the first clock signal supply line CL1. And, a backward isolation transistor Q_in3b is provided so as to isolate the signal-storage state of the signal-storage state of the (n-1)-th bit-level cell M_in-1 in the (n-1)-th memory unit U_n-1 from the n-th bit-level cell M_in in the n-th memory unit U_n, the backward isolation transistor Q_in3b transfers backward a signal from the n-th bit-level cell M_in to the (n-1)-th bit-level cell M_in-1 at a required timing determined by a clock signal, which is supplied through the second clock signal supply line CL2. Then, a sequence of the forward isolation transistors Q_in3f (i=1 to m) arrayed in parallel with the memory units U_n-1 and U_n transfers the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL1 so that the information of byte size or word size can march along the forward direction, pari passu. And, a sequence of the backward isolation transistors Q_in3b (i=1 to m) arrayed in parallel with the memory units U_n-1 and U_n transfers the information of byte size or word size, controlled by the clock signal supplied through the clock signal supply line CL2 so that the information of byte size or word size can march along the backward direction, pari passu.

In the configuration of the bidirectional marching main memory illustrated in Figs. 32 and 33, the forward capacitor C_ijf and the backward capacitor C_ij shall preferably be merged into a single common capacitor so as to implement random access mode with high locality. Fig. 34 illustrates an array of i-th row of the m * n matrix (here, "m" is an integer determined by word size) in a gate-level representation of the bidirectional marching main memory 31, which can achieve the random access mode in the bidirectional behavior illustrated in Figs. 31 (a)-(c).

As illustrated in Fig. 34, two kinds of marching AND-gates are assigned to each of the cells M_i1, M_i2, M_i3,........., M_i,(n-1), M_i,n on the i-th row so as to establish a bidirectional transfer of information/data with random access mode. The bidirectional marching main memory 31 stores the information/data of bit level in each of cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,n and transfers bi-directionally the information/data synchronously with the clock signal, step by step in the forward direction and/or reverse direction (backward direction) between a first I/O selector 512 and a second I/O selector 513.

In the gate-level representation of cell array implementing the marching main memory 31 illustrated in Fig. 34, a first bit-level cell M_i1 allocated at the leftmost side on i-th row and connected to first I/O selector 512 encompasses a common capacitor C_i1configured to store the information/data, and a forward marching AND-gate G_i1fhaving one input terminal connected to the common capacitor C_i1, the other input supplied with the first clock signal supply line CL1, and an output terminal connected to one input terminal of the next forward marching AND-gate G_(i+1)1f assigned to the adjacent second bit-level cell M_(i+1)1 on the i-th row, and a backward marching AND-gate G_i1bhaving one input terminal connected to the common capacitor C_i1, the other input supplied with the second clock signal supply line CL2, and an output terminal connected to the first I/O selector 512.

The first clock signal supply line CL1, configured to drive the forward data-stream, and the second clock signal supply line CL2, configured to drive the backward data-stream, are respectively selected by a clock selector 511, and each of the first clock signal supply line CL1 and the second clock signal supply line CL2 has logical values of "1" and "0". When the logical values of "1" of the first clock signal supply line CL1 is fed to the other input terminal of the forward marching AND-gate G_i1, the information/data stored in the common capacitor C_i1 is transferred to a common capacitor C_i2, assigned to the adjacent second bit-level cell M_i2, and the common capacitor C_i2stores the information/data.

The second bit-level cell M_i2 on the i-th row of the bidirectional marching main memory 31 encompasses the common capacitor C_i2configured to store the information/data, a forward marching AND-gate G_i2f, which has one input terminal connected to the common capacitor C_i2, the other input supplied with the first clock signal supply line CL1, and an output terminal connected to one input terminal of the next forward marching AND-gate G₁₃assigned to the adjacent third bit-level cell M_i3 on the i-th row, and the backward marching AND-gate G_i2bhaving one input terminal connected to the common capacitor C_i2, the other input supplied with the second clock signal supply line CL2, and an output terminal connected to one input terminal of the preceding backward marching AND-gate G_i1.

Similarly the third bit-level cell M_i3 on the i-th row encompasses a common capacitor C_i3configured to store the information/data, a forward marching AND-gate G_i3fhaving one input terminal connected to the common capacitor C_i3, the other input supplied with the first clock signal supply line CL1, and an output terminal connected to one input terminal of the next forward marching AND-gate assigned to the adjacent fourth cell, although the illustration of the fourth cellis omitted, and an backward marching AND-gate G_i3bhaving one input terminal connected to the common capacitor C_i3, the other input supplied with the second clock signal supply line CL2, and an output terminal connected to one input terminal of the preceding backward marching AND-gate G_i2b assigned to the adjacent second bit-level cell M_i2. Therefore, when the logical values of "1" of the first clock signal supply line CL1 is fed to the other input terminal of the forward marching AND-gate G_i2f, the information/data stored in the common capacitor C_i2 is transferred to the common capacitor C_i3, assigned to the third bit-level cell M_i3, and the common capacitor C_i3stores the information/data, and when the logical values of "1" of the first clock signal supply line CL1 is fed to the other input terminal of the forward marching AND-gate G_i3f, the information/data stored in the common capacitor C_i3 is transferred to the capacitor, assigned to the fourth cell.

Furthermore, an (n-1)-th bit-level cell M_i,(n-1) on the i-th row encompasses a common capacitor C_i,(n-1), configured to store the information/data, and a forward marching AND-gate G_i,(n-1)fhaving one input terminal connected to the common capacitor C_i,(n-1), the other input supplied with the first clock signal supply line CL1, and an output terminal connected to one input terminal of the next forward marching AND-gate G_i,nf assigned to the adjacent n-th bit-level cell M_i,n, which is allocated at the rightmost side on the i-th row and connected to the second I/O selector 513, and an backward marching AND-gate G_i,(n-1)b, which has one input terminal connected to the common capacitor C_i,(n-1), the other input supplied with the second clock signal supply line CL2, and an output terminal connected to one input terminal of the preceding backward marching AND-gate G_i,(n-2)b assigned to the adjacent third bit-level cell M_i,(n-2)b(illustration is omitted).

Finally, an n-th bit-level cell M_i,n allocated at the rightmost side on the i-th row and connected to the second I/O selector 513 encompasses a common capacitor C_i,nconfigured to store the information/data, a backward marching AND-gate G_inbhaving one input terminal connected to the common capacitor C_in, the other input terminal configured to be supplied with the second clock signal supply line CL2, and an output terminal connected to one input terminal of the preceding backward marching AND-gate G_i(n-1)b assigned to the adjacent (n-1)-th bit-level cell M_i,n-1 on the i-th row, and a forward marching AND-gate G_i,nfhaving one input terminal connected to the common capacitor C_i,n, the other input terminal configured to be supplied with the first clock signal supply line CL1, and an output terminal connected to the second I/O selector 513.

When the logical values of "1" of the second clock signal supply line CL2 is fed to the other input terminal of the backward marching AND-gate G_inb, the information/data stored in the common capacitor C_in is transferred to a common capacitor C_i,(n-1), assigned to the adjacent (n-1)-th bit-level cell M_i,(n-1) on the i-th row, and the common capacitor C_i,(n-1)stores the information/data. Then, when the logical values of "1" of the second clock signal supply line CL2 is fed to the other input terminal of the backward marching AND-gate G_i3b, the information/data stored in the common capacitor C_i3 is transferred to the common capacitor C_i2, assigned to the second bit-level cell M_i2, and the common capacitor C_i2stores the information/data. Furthermore, when the logical values of "1" of the second clock signal supply line CL2 is fed to the other input terminal of the backward marching AND-gate G_i2b, the information/data stored in the common capacitor C_i2 is transferred to the common capacitor C_i1, assigned to the second bit-level cell M_i1, and the common capacitor C_i1stores the information/data, and when the logical values of "1" of the second clock signal supply line CL2 is fed to the other input terminal of the backward marching AND-gate G_i1b, the information/data stored in the common capacitor C_i1 is transferred to the first I/O selector 512.

Therefore, each of the cells M_i1, M_i2, M_i3,........., M_i,(n-1), M_i,n on the i-th row of the bidirectional marching main memory stores the information/data, and transfers bi-directionally the information/data, synchronously with the clock signals supplied respectively from the first clock signal supply line CL1 and the second clock signal supply line CL2, step by step, between the first I/O selector 512 and the second I/O selector 513. Because each of the cells M_i1, M_i2, M_i3,........., M_i,n-1, M_i,nis assigned in memory unit U₁, U₂, U₃,........., U_n-1, U_n, respectively, and the memory units U₂, U₃,........., U_n-1, U_n stores information of byte size or word size by the sequence of bit-level cells arrayed in the memory unit U₂, U₃,........., U_n-1, U_n, respectively, the bidirectional marching main memory 31 illustrated in Fig.34 stores the information/data of byte size or word size in each of cells U₁, U₂, U₃,........., U_n-1, U_n and transfers bi-directionally the information/data of byte size or word size synchronously with the clock signal, pari passu, in the forward direction and/or reverse direction (backward direction) between a first I/O selector 512 and a second I/O selector 513, so as to provide the processor 11 with the stored information/data of byte size or word size actively and sequentially so that the ALU 112 can execute the arithmetic and logic operations with the stored information/data.

(POSITION POINTING STRATEGY)
Fig. 35(a) illustrates a bidirectional transferring mode of instructions in a one-dimensional marching main memory adjacent to a processor, where the instructions moves toward the processor, and moves from / to the next memory. Fig. 35(b) illustrates a bidirectional transferring mode of scalar data in a one-dimensional marching main memory adjacent to an ALU 112, the scalar data moves toward the ALU and moves from / to the next memory. Fig. 35(c) illustrates a uni-directional transferring mode of vector/streaming data in a one-dimensional marching main memory adjacent to a pipeline 117, which will be explained in the following third embodiment, the vector/streaming data moves toward the pipeline 117, and moves from the next memory.

The marching main memory 31 used in the computer system pertaining to the first embodiment uses positioning to identify the starting point and ending point of a set of successive memory units U₁, U₂, U₃,........., U_n-1, U_n in vector/streaming data. On the other hand, for programs and scalar data, each item must have a position index similar to conventional address. Fig. 36(a) illustrates a configuration of conventional main memory, in which every memory units U₁, U₂, U₃,........., U_n-1, U_n in are labeled by addresses A₁, A₂, A₃,........., A_n-1, A_n, Fig. 36(b) illustrates a configuration of one-dimensional marching main memory, in which the positioning of individual memory unit U₁, U₂, U₃,........., U_n-1, U_n is not always necessary, but the positioning of individual memory unit U₁, U₂, U₃,........., U_n-1, U_n is at least necessary to identify the starting point and ending point of a set of successive memory units in vector/streaming data.
Fig. 37(a) illustrates an inner configuration of present one-dimensional marching main memory, in which the position indexes like existing addresses are not necessary for scalar instruction I_s, but the positioning of individual memory unit is at least necessary to identify the starting point and ending point of a set of successive memory units in vector instruction I_v, as indicated by hatched circle. Fig. 37(b) illustrates an inner configuration of present one-dimensional marching main memory, in which the position indexes are not necessary for scalar data "b" and "a". However, as illustrated in Fig. 37(c), position indexes are at least necessary to identify the starting point and ending point of a set of successive memory units in vector/streaming data "o", "p", "q", "r", "s", "t",.......... as indicated by hatched circle.

In a marching memory family, which includes a marching-instruction register file 22a and a marching-data register file 22b connected to the ALU 112, which will be explained in the following second embodiment, and a marching-instruction cache memory 21a and a marching-data cache memory 21b, which will be explained in the following third embodiment, in addition to the marching main memory 31 used in the computer system pertaining to the first embodiment of the present invention, the relation between the main memory, the register file and cache memory has their own position pointing strategy based on the property of locality of reference.

Fig. 38(a) illustrates schematically an example of an overall configuration of present marching main memory implemented by a plurality of pages P_i-1,j-1, P_i,j-1, P_i+1,j-1, P_i+2,j-1, P_i-1,j, P_i,j, P_i+1,j, P_i+2,j for vector/streaming data case. Fig. 38(b) illustrates schematically an example of a configuration of the hatched page P_i,j, which is implemented by a plurality of files F₁, F₂, F₃, F₄ for vector/streaming data case, and each of the pages P_i-1,j-1, P_i,j-1, P_i+1,j-1, P_i+2,j-1, P_i-1,j, P_i,j, P_i+1,j, P_i+2,j can be used for

marching cache memories

21a and 21b in the third embodiment. Fig. 38(c) illustrates schematically an example of a configuration of the hatched file F₃, each of the files F₁, F₂, F₃, F₄ is implemented by a plurality of memory units U₁, U₂, U₃,........., U_n-1, U_n for vector/streaming data case, and each of the file files F₁, F₂, F₃, F₄ can be used for marching

register files

22a and 22b in the second embodiment.

Similarly, Fig. 39(a) illustrates schematically an example of an overall configuration of present marching main memory implemented by a plurality of pages P_r-1,s-1, P_r,s-1, P_r+1,s-1, P_r+2,s-1, P_r-1,s, P_r,s, P_r+1,s, P_r+2,s for programs/scalar data case, where each pages has its own position index as an address. Fig. 39(b) illustrates schematically an example of a configuration of the hatched page P_r-1,s and the driving positions of the page P_r-1,s, using digits in the binary system, each of the page P_r-1,s-1, P_r,s-1, P_r+1,s-1, P_r+2,s-1, P_r-1,s, P_r,s, P_r+1,s, P_r+2,s is implemented by a plurality of files F₁, F₂, F₃, F₄ for programs/scalar data case. Each of the page P_r-1,s-1, P_r,s-1, P_r+1,s-1, P_r+2,s-1, P_r-1,s, P_r,s, P_r+1,s, P_r+2,s can be used for

marching cache memories

21a and 21b in the third embodiment, where each of the files F₁, F₂, F₃, F₄ has its own position index as address. Fig. 39(c) illustrates schematically an example of a configuration of the hatched file F₃ and the driving positions of the file F₃, using

digits

0, 1, 2, 3 in the binary system, each of the files F₁, F₂, F₃, F₄ is implemented by a plurality of memory units U₁, U₂, U₃,........., U_n, U_n+1 , U_n+2 , U_n+3 , U_n+4 , U_n+5 for programs/scalar data case. Each of the files F₁, F₂, F₃, F₄ can be used for a

marching register files

22a and 22b in the second embodiment, where each memory units U₁, U₂, U₃,........., U_n, U_n+1 , U_n+2 , U_n+3 , U_n+4 , U_n+5 has its own position index n+4, n+3, n+2,........., 5, 4, 3, 2, 1, 0 as address. Fig. 39(c) represents position pointing strategy for all of the cases by digits in the binary system.

As illustrated in Fig. 39(c), the n binary digits identify a single memory unit among 2ⁿ memory units, respectively, in a memory structure having an equivalent size corresponding to the size of a marching register file. And, as illustrated in Fig. 39(b), the structure of one page has an equivalent size corresponding to the size of a marching cache memory, which is represented by two digits which identify four files F₁, F₂, F₃, F₄, while the structure of one marching main memory is represented by three digits which identify eight pages P_r-1,s-1, P_r,s-1, P_r+1,s-1, P_r+2,s-1, P_r-1,s, P_r,s, P_r+1,s, P_r+2,s in the marching main memory as illustrated in Fig. 39(a).

(SPEED/CAPABILITY)
The speed gap between memory access time and the CPU cycle time in a conventional computer system is, for example, 1:100. However, the speed of the marching memory access time is equal to the CPU cycle time in the computer system of the first embodiment. Fig. 40 compares the speed/capability of the conventional computer system without cache with that of the marching main memory 31, configured to be used in the computer system pertaining to the first embodiment of the present invention. That is, Fig. 40(b) illustrates schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀, and compares with the speed/capability of the existing memory illustrated in Fig. 40(a). We can also support 99 additional simultaneous memory units of the marching main memory 31, in the condition that we have necessary processing units to use the data from the marching main memory 31. Therefore, one memory unit time T_mue in the conventional computer system is estimated to be equal to one hundred of the memory unit streaming time T_mus of the marching main memory 31 pertaining to the first embodiment of the present invention.

And, Fig. 41 compares the speed/capability of the worst case of the existing memory for scalar data or program instructions with that of the marching main memory 31, configured to be used in the computer system pertaining to the first embodiment of the present invention. That is, the hatched portion of Fig. 41(b) illustrates schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀, and compares with the speed/capability of the worst case of the existing memory illustrated in Fig. 41(a). In the worst case, we can read out 99 memory units of the marching main memory 31, but they are not available due to a scalar program's requirement.

Further, Fig. 42 compares the speed/capability of the typical case of the existing memory for scalar data or program instructions with that of the marching main memory 31, configured to be used in the computer system pertaining to the first embodiment of the present invention. That is, Fig. 42(b) illustrates schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀, and compares with the speed/capability of the typical case of the existing memory illustrated in Fig. 42(a). In the typical case, we can read out 99 memory units but only several memory units are available, as illustrated by hatched memory units in the existing memory, by speculative data preparation in a scalar program.

Fig. 43 compares the speed/capability of the typical case of the existing memory for scalar data case with that of the marching main memory 31, configured to be used in the computer system pertaining to the first embodiment of the present invention. That is, Fig. 43(b) illustrates schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀, and compares with the speed/capability of the existing memory illustrated in Fig. 43(a). Similar to the case illustrated in Figs. 34 (a)-(b), in the typical case, we can read out 99 memory units but only several memory units are available, as illustrated by hatched memory units in the existing memory, by speculative data preparation in a scalar data or program instructions in multi-thread parallel processing.

Fig. 44 compares the speed/capability of the best case of the existing memory for streaming data, vector data or program instructions case with that of the marching main memory 31, configured to be used in the computer system pertaining to the first embodiment of the present invention. That is, Fig. 44(b) illustrates schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀,, and compares with the speed/capability of the best case of the existing memory illustrated in Fig. 44(a). In the best case, we can understand that one hundred memory units of the marching main memory 31 are usable for streaming data and data parallel.

(TWO-DIMENSIONAL MARCHING MAIN MEMORY)
The memory units can be arranged two-dimensionally on a chip as illustrated in Figs. 45-51 so that various mode of operation can be achieved without a switch/network. According to the two-dimensional marching main memory 31 of the first embodiment illustrated in Figs. 45-51, the memory units U₁₁, U₁₂, U₁₃,........., U_{1, v-1}, U_1v; U₂₂, U₂₂, U₂₃,........., U_{2, v-2}, U_2v; ..........; U_u1, U_u2, U_u3,........., U_{u, v-1}, U_uv are not required of the refreshment, because all of the memory units U₁₁, U₁₂, U₁₃,........., U_{1, v-1}, U_1v; U₂₂, U₂₂, U₂₃,........., U_{2, v-2}, U_2v; ..........; U_u1, U_u2, U_u3,........., U_{u, v-1}, U_uv are usually refreshed automatically due to the information-moving scheme (information-marching scheme). And then addressing to each of memory units U₁₁, U₁₂, U₁₃,........., U_{1, v-1}, U_1v; U₂₂, U₂₂, U₂₃,........., U_{2, v-2}, U_2v; ..........; U_u1, U_u2, U_u3,........., U_{u, v-1}, U_uv disappears and required information is heading for its destination unit connected to the edge of the memory. The mechanism of accessing the two-dimensional marching main memory 31 of the first embodiment is truly alternative to existing memory schemes that are starting from the addressing mode to read/write information in the conventional computer system. Therefore, according to the two-dimensional marching main memory 31 of the first embodiment, the memory-accessing process without addressing mode in the computer system of the first embodiment is quite simpler than existing memory schemes of the conventional computer system.

(ENERGY CONSUMPTION)
To clarify the improvement of architecture, design and implementation of the computer system pertaining to the first embodiment of the present invention, the improvement in energy consumption will be explained. Fig. 52(a) shows that the energy consumption in microprocessors can be decomposed into static power consumption and dynamic power consumption. In the dynamic power consumption illustrated in Fig. 52(a), net and overhead of the power consumption are outstandingly illustrated in Fig. 52(b). As illustrated in Fig. 52(c), only the net energy portions are practically necessary to operate a given job in a computer system, so that these pure energy parts make least energy consumption to perform the computer system. This means the shortest processing time is achieved by the net energy consumed illustrated in Fig. 52(c).

Even though some efforts are introduced into architecting, designing and implementing processors, there are bottlenecks in the conventional architecture as illustrated in Fig. 1. In the conventional architecture, there are various issues in the von Neumann computer, as follows:

1) Programs are stored like data in memory;
2) All processing is basically sequential in a uni-processor;
3) The operation of programs is the sequential execution of instructions;
4) Vector data is sequentially processed by the CPU with vector instructions;
5) Streaming data is sequentially processed with threads;
6) Programs then threads are arranged sequentially;
7) Data parallel consists of an arrangement of data as a vector: and
8) Streaming data is a flow of data

From the properties of a conventional computer, we conclude that storage of programs and data is in a fashion of basically sequentially arranged ones. This fact means the regular arrangement of instructions exists in a program and the corresponding data. .

In the computer system pertaining to the first embodiment of the present invention illustrated in Fig.2, the access of instructions in the marching main memory 31 is not necessary, because instructions are actively accessed by themselves to processor 11. Similarly, the access of data in the marching main memory 31 is not necessary, because data are actively accessed by themselves to processor 11.

Fig. 53 shows an actual energy consumption distribution over a processor including registers and caches in the conventional architecture, estimated by William J. Dally, et al., in "Efficient Embedded Computing", Computer, vol. 41, no. 7, 2008, pp. 27-32. In Fig. 53, an estimation of the power consumption distribution on only the whole chip, except for wires between chips is disclosed. By Dally, et al, the instruction supply power consumption is estimated to be 42%, the data supply power consumption is estimated to be 28%, the clock and control logic power consumption is estimated to be 24%, and the arithmetic power consumption is estimated to be 6%. Therefore, we can understand that instruction supply and data supply power consumptions are relatively larger than of the clock/ control logic power consumption and the arithmetic power consumption, which is ascribable to the inefficiency of cache/register accessing with lots of wires and some software overhead due to access ways of these caches and registers in addition to non-refreshment of all the memories, caches and registers.

Because the ratio of the instruction supply power consumption to the data supply power consumption is 3:2, and the ratio of the clock and control logic power consumption to the arithmetic power consumption is 4:1, in accordance with the computer system pertaining to the first embodiment of the present invention illustrated in Fig.2, we can reduce easily the data supply power consumption to 20% by using the marching main memory 31 at least partly so that the instruction supply power consumption becomes 30%, while we can increase the arithmetic power consumption to 10% so that the clock and control logic power consumption become 40%, which means that the sum of the instruction supply power consumption and the data supply power consumption can be made 50%, and the sum of the clock and control logic power consumption and the arithmetic power consumption can be made 50%.

If we reduce the data supply power consumption to 10%, the instruction supply power consumption becomes 15%, and if we increase the arithmetic power consumption to 15%, the clock and control logic power consumption will become 60%, which means that the sum of the instruction supply power consumption and the data supply power consumption can be made 35%, while the sum of the clock and control logic power consumption and the arithmetic power consumption can be made 75%.

The conventional computer system dissipates energy as illustrated in the Fig. 54(a) with a relatively large average active time for addressing and read/writing memory units, accompanied by wire delay time, while the present computer system dissipates smaller energy as illustrated in the Fig. 54(b), because the present computer system has a shorter average active smooth time through marching memory, and we could process the same data faster than the conventional computer system with less energy.

--SECOND EMBODIMENT--
As illustrated in Fig. 55, a computer system pertaining to a second embodiment of the present invention encompasses a processor 11 and a marching main memory 31. The processor 11 includes a control unit 111 having a clock generator 113 configured to generate a clock signal, an arithmetic logic unit (ALU) 112 configured to execute arithmetic and logic operations synchronized with the clock signal, a marching-instruction register file (RF) 22a connected to the control unit 111 and a marching-data register file (RF) 22b connected to the ALU 112.

Although the illustration is omitted, very similar to the marching main memory 31 illustrated in Figs. 3-24, 25(a), 25(b), 26 and 45-51, the marching-instruction register file 22a has an array of instruction register units, instruction-register input terminals of the third array configured to receive the stored instruction from the marching main memory 31, and instruction-register output terminals of the third array, configured to store instruction in each of instruction register units and to transfer successively and periodically the stored instruction in each of instruction register units to an adjacent instruction register unit being synchronized with the clock signal from the instruction register units adjacent to the instruction-register input terminals toward the instruction register units adjacent to the instruction-register output terminals, so as to provide actively and sequentially instruction implemented by the stored instruction to the control unit 111 through the instruction-register output terminals so that the control unit 111 can execute operations with the instruction.

Further similar to the marching main memory 31 illustrated in Figs. 3-24, 25(a), 25(b), 26 and 45-51, the marching-data register file 22b has an array of data register units, data-register input terminals of the fourth array configured to receive the stored data from the marching main memory 31, and data-register output terminals of the fourth array, configured to store data in each of data register units and to transfer successively and periodically the stored data in each of data register units to an adjacent data register unit being synchronized with the clock signal from the data register units adjacent to the data-register input terminals toward the data register units adjacent to the data-register output terminals, so as to provide actively and sequentially the data to the ALU 112 through the data-register output terminals so that the ALU 112 can execute operations with the data, although the detailed illustration of, the marching-data register file 22b is omitted,.

As illustrated in Fig. 55, a portion of the marching main memory 31 and the marching-instruction register file 22a are electrically connected by a plurality of joint members 54, and remaining portion of the marching main memory 31 and the marching-data register file 22b are electrically connected by another plurality of joint members 54.

The resultant data of the processing in the ALU 112 are sent out to the marching-data register file 22b. Therefore, as represented by bidirectional arrow PHI(Greek-letter)₂₄, data are transferred bi-directionally between the marching-data register file 22b and the ALU 112. Furthermore, the data stored in the marching-data register file 22b are sent out to the marching main memory 31 through the joint members 54. Therefore, as represented by bidirectional arrow PHI₂₃, data are transferred bi-directionally between the marching main memory 31 and the marching-data register file 22b through the joint members 54.

On the contrary, as represented by uni-directional arrows ETA(Greek-letter)₂₂ and ETA₂₃, as to the instructions movement, there is only one way of instruction-flow from the marching main memory 31 to the marching-instruction register file 22a, and from the marching-instruction register file 22a to the control unit 111.

In the computer system of the second embodiment illustrated in Fig. 55, there are no buses consisting of the data bus and address bus because the whole computer system has no wires even in any data exchange between the marching main memory 31 and the marching-instruction register file 22a, between the marching main memory 31 and the marching-data register file 22b, between the marching-instruction register file 22a and the control unit 111 and between the marching-data register file 22band the ALU 112, while the wires or the buses implement the bottleneck in the conventional computer system. As there are no global wires, which generate time delay and stray capacitances between these wires, the computer system of the second embodiment can achieve much higher processing speed and lower power consumption.

Since other functions, configurations, and ways of operation of the computer system pertaining to the second embodiment are substantially similar to the functions, configurations, way of operation already explained in the first embodiment, overlapping or redundant description may be omitted.

--THIRD EMBODIMENT--
As illustrated in Fig. 56, a computer system pertaining to a third embodiment of the present invention encompasses a processor 11, a marching-cache memory (21a, 21b) and a marching main memory 31. Similar to the second embodiment, the processor 11 includes a control unit 111 having a clock generator 113 configured to generate a clock signal, an arithmetic logic unit (ALU) 112 configured to execute arithmetic and logic operations synchronized with the clock signal, a marching-instruction register file (RF) 22a connected to the control unit 111 and a marching-data register file (RF) 22b connected to the ALU 112.

The marching-cache memory (21a, 21b) embraces a marching-instruction cache memory 21a and a marching-data cache memory 21b. Although the illustration is omitted, very similar to the marching main memory 31 illustrated in Figs. 3-24, 25(a), 25(b), 26 and 45-51, each of the marching-instruction cache memory 21a and the marching-data cache memory21b has an array of cache memory units at locations corresponding to a unit of information, cache input terminals of the array configured to receive the stored information from the marching main memory 31, and cache output terminals of the array, configured to store information in each of cache memory units and to transfer, synchronously with the clock signal, step by step, the information each to an adjacent cache memory unit, so as to provide actively and sequentially the stored information to the processor 11 so that the ALU 112 can execute the arithmetic and logic operations with the stored information.

As illustrated in Fig. 56, a portion of the marching main memory 31 and the marching-instruction cache memory 21a are electrically connected by a plurality of joint members 52, and remaining portion of the marching main memory 31 and the marching-data cache memory 21b are electrically connected by another plurality of joint members 52. Furthermore, the marching-instruction cache memory 21a and the marching-instruction register file 22a are electrically connected by a plurality of joint members 51, and the marching-data cache memory 21b and the marching-data register file 22b are electrically connected by another plurality of joint members 51.

The resultant data of the processing in the ALU 112 are sent out to the marching-data register file 22b, and, as represented by bidirectional arrow PHI(Greek-letter)₃₄, data are transferred bi-directionally between the marching-data register file 22b and the ALU 112. Furthermore, the data stored in the marching-data register file 22b are sent out to the marching-data cache memory 21b through the joint members 51, and, as represented by bidirectional arrow PHI₃₃, data are transferred bi-directionally between the marching-data cache memory 21b and the marching-data register file 22b through the joint members 51. Furthermore, the data stored in the marching-data cache memory 21b are sent out to the marching main memory 31 through the joint members 52, and, as represented by bidirectional arrow PHI₃₂, data are transferred bi-directionally between the marching main memory 31 and the marching-data cache memory 21b through the joint members 52.

On the contrary, as represented by uni-directional arrows ETA(Greek-letter)₃₁, eta₃₂ and eta₃₃, as to the instructions movement, there is only one way of instruction-flow from the marching main memory 31 to the marching-instruction cache memory 21a, from the marching-instruction cache memory 21a to the marching-instruction register file 22a, and from the marching-instruction register file 22a to the control unit 111.

In the computer system of the third embodiment illustrated in Fig. 56, there are no buses consisting of the data bus and address bus because the whole computer system has no global wires even in any data exchange between the marching main memory 31 and the marching-instruction cache memory 21a, between the marching-instruction cache memory 21a and the marching-instruction register file 22a, between the marching main memory 31 and the marching-data cache memory 21b, between the marching-data cache memory 21b and the marching-data register file 22b, between the marching-instruction register file 22a and the control unit 111 and between the marching-data register file 22band the ALU 112, while the wires or the buses implement the bottleneck in the conventional computer system. As there are no global wires, which generate time delay and stray capacitances between these wires, the computer system of the third embodiment can achieve much higher processing speed and lower power consumption.

Since other functions, configurations, way of operation of the computer system pertaining to the third embodiment are substantially similar to the functions, configurations, way of operation already explained in the first and second embodiments, overlapping or redundant description may be omitted.
As illustrated in Fig. 57(a), the ALU 112 in the computer system of the third embodiment may includes a plurality of arithmetic pipelines P₁, P₂, P₃,........., P_n configured to receive the stored information through marching register units R₁₁, R₁₂, R₁₃,........., R_1n; R₂₂, R₂₂, R₂₃,........., R_2n, in which data move in parallel with the alignment direction of the arithmetic pipelines P₁, P₂, P₃,........., P_n. In case that vector data are stored, marching-vector register units R₁₁, R₁₂, R₁₃,........., R_1n; R₂₂, R₂₂, R₂₃,........., R_2n can be used.

Furthermore, as illustrated in Fig. 57(b), a plurality of marching cache units C₁₁, C₁₂, C₁₃,........., C_1n; C₂₁, C₂₂, C₂₃,........., C_2n; C₃₁, C₃₂, C₃₃,........., C_3n can be aligned in parallel.

As illustrated in Fig. 58, the ALU 112 in the computer system of the third embodiment may include a single processor core 116, and as represented by cross-directional arrows, the information can moves from the marching-cache memory 21 to the marching-register file 22, and from the marching-register file 22 to the processor core 116. The resultant data of the processing in the processor core 116 are sent out to the marching-register file 22 so that data are transferred bi-directionally between the marching-register file 22 and the processor core 116. Furthermore, the data stored in the marching-register file 22 are sent out to the marching-cache memory 21 so that data are transferred bi-directionally between the marching-cache memory 21 and the marching-register file 22. In case of instructions movement, there is no flow along the opposite direction of the information to be processed.

As illustrated in Fig. 59, the ALU 112 in the computer system of the third embodiment may include a single arithmetic pipeline 117, and as represented by cross-directional arrows, the information can moves from the marching-cache memory 21 to the marching-vector register file 22v, and from the marching-vector register file 22v to the arithmetic pipeline 117. The resultant data of the processing in the arithmetic pipeline 117 are sent out to the marching-vector register file 22v so that data are transferred bi-directionally between the marching-vector register file 22v and the arithmetic pipeline 117. Furthermore, the data stored in the marching-vector register file 22v are sent out to the marching-cache memory 21 so that data are transferred bi-directionally between the marching-cache memory 21 and the marching-vector register file 22v. In case of instructions movement, there is no flow along the opposite direction of the information to be processed.

As illustrated in Fig. 60, the ALU 112 in the computer system of the third embodiment may include a plurality of

processor cores

116_-1, 116_-2, 116_-3, 116_-4,........., 116_-m, and as represented by cross-directional arrows, the information can moves from the marching-cache memory 21 to the marching-register file 22, and from the marching-register file 22 to the

processor cores

116_-1, 116_-2, 116_-3, 116_-4,........., 116_-m. The resultant data of the processing in the

processor cores

116_-1, 116_-2, 116_-3, 116_-4,........., 116_-m are sent out to the marching-register file 22 so that data are transferred bi-directionally between the marching-register file 22 and the

processor cores

116_-1, 116_-2, 116_-3, 116_-4,........., 116_-m. Furthermore, the data stored in the marching-register file 22 are sent out to the marching-cache memory 21 so that data are transferred bi-directionally between the marching-cache memory 21 and the marching-register file 22. In case of instructions movement, there is no flow along the opposite direction of the information to be processed.

As illustrated in Fig. 61, the ALU 112 in the computer system of the third embodiment may include a plurality of

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m, and as represented by cross-directional arrows, the information can moves from the marching-cache memory 21 to the marching-vector register file 22v, and from the marching-vector register file 22v to the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m. The resultant data of the processing in the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m are sent out to the marching-vector register file 22v so that data are transferred bi-directionally between the marching-vector register file 22v and the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m. Furthermore, the data stored in the marching-vector register file 22v are sent out to the marching-cache memory 21 so that data are transferred bi-directionally between the marching-cache memory 21 and the marching-vector register file 22v. In case of instructions movement, there is no flow along the opposite direction of the information to be processed.

As illustrated in Fig. 62(b), the ALU 112 in the computer system of the third embodiment may include a plurality of

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m, and a plurality of

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m are electrically connected to the marching main memory 31. Here, a first marching-vector register file 22v_-1 is connected to the first marching-cache memory 21_-1, and a first arithmetic pipeline 117_-1is connected to the first marching-vector register file 22v_-1. And, a second marching-vector register file 22v_-2 is connected to the second marching-cache memory 21_-2, and a second arithmetic pipelines 117_-2is connected to the second marching-vector register file 22v_-2; a third marching-vector register file 22v_-3 is connected to the third marching-cache memory 21_-3, and a third arithmetic pipelines 117_-3is connected to the third marching-vector register file 22v_-3;........; and a m-th marching-vector register file 22v_-m is connected to the m-th marching-cache memory 21_-m, and a m-th arithmetic pipelines 117_-mis connected to the m-th marching-vector register file 22v_-m.

The information moves from the marching main memory 31 to the

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m in parallel, from

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m to the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m in parallel, and from the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m to the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m in parallel. The resultant data of the processing in the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m are sent out to the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m so that data are transferred bi-directionally between the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m and the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m. Furthermore, the data stored in the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m are sent out to the

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m so that data are transferred bi-directionally between the

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m and the marching-

vector register files

22v_-1, 22v_-2, 22v_-3, 22v_-4,........., 22v_-m, and the data stored in the

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m are sent out to the marching main memory 31 so that data are transferred bi-directionally between the marching main memory 31 and the

marching cache memories

21_-1, 21_-2, 21_-3, 21_-4,........., 21_-m. In case of instructions movement, there is no flow along the opposite direction of the information to be processed.

On the contrary, as illustrated Fig. 62(a), in the ALU 112 of the conventional computer system including a plurality of

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m, a plurality of

conventional cache memories

321_-1, 321_-2, 321_-3, 321_-4,........., 321_-m are electrically connected to the conventional main memory 331 through wires and/or buses which implement von Neumann bottleneck 325. Then, information moves from the conventional main memory 331 to the

conventional cache memories

321_-1, 321_-2, 321_-3, 321_-4,........., 321_-m in parallel through von Neumann bottleneck 325, from

conventional cache memories

321_-1, 321_-2, 321_-3, 321_-4,........., 321_-m to the conventional-vector register files (RFs) 322v_-1, 322v_-2, 322v_-3, 322v_-4,........., 322v_-m in parallel, and from the conventional-

vector register files

322v_-1, 322v_-2, 322v_-3, 322v_-4,........., 322v_-m to the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m in parallel.

In the computer system of the third embodiment illustrated in Fig. 62(b), there are no buses consisting of the data bus and address bus because the whole system has no global wires even in any data exchange between the

arithmetic pipelines

117_-1, 117_-2, 117_-3, 117_-4,........., 117_-m and the marching main memory 31, while the wires or the buses implement the bottleneck in the conventional computer system as illustrated in Fig. 62(a). As there are no global wires, which generate time delay and stray capacitances between these wires, the computer system illustrated in Fig. 62(b) can achieve much higher processing speed and lower power consumption.

--FOURTH EMBODIMENT--
As illustrated in Fig. 63, a computer system of a fourth embodiment encompasses a conventional main memory 31s, a mother marching main memory 31_-0 connected to the conventional main memory 31s, and a plurality of

processing units

12_-1, 12_-2, 12_-3,........, configured to communicate with mother marching main memory 31_-0 so as to implement a high performance computing (HPC) system, which can be used for graphics processing unit (GPU)-based general-purpose computing. Although the illustration is omitted, the HPC system of the fourth embodiment further includes a control unit 111 having a clock generator 113 configured to generate a clock signal, and a field programmable gate array (FPGA) configured to switch-control operations of the plurality of

processing units

12_-1, 12_-2, 12_-3,.........., optimizing the flow of crunching calculations by running parallel, constructing to help manage and organize bandwidth consumption. FPGA is, in essence, a computer chip that can rewire itself for a given task. FPGA can be programmed with hardware description languages such as VHDL or Verilog.

The first processing unit 12_-1 encompasses a first branched-marching main memory 31_-1, a plurality of first

marching cache memories

21_-11, 21_-12,........., 21_-1p electrically connected respectively to the first branched-marching main memory 31_-1, a plurality of first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p electrically connected respectively to the first

marching cache memories

21_-11, 21_-12,........., 21_-1p, a plurality of first

arithmetic pipelines

117_-11, 117_-12,........., 117_-1p electrically connected respectively to the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p.

Similar to the configurations illustrated in Figs. 3-24, 25(a), 25(b), 26 and 45-51 etc., because each of the mother marching main memory 31_-0, the first branched-marching main memory 31_-1, the first

marching cache memories

21_-11, 21_-12,........., 21_-1p, and the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p encompasses an array of memory units, input terminals of the array and output terminals of the array, configured to store information in each of memory units and to transfer synchronously with the clock signal, step by step, from a side of input terminals toward the output terminals.

Because the operations of the mother marching main memory 31_-0, the first branched-marching main memory 31_-1, the first

marching cache memories

21_-11, 21_-12,........., 21_-1p, and the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p are controlled by FPGA, the information moves from the mother marching main memory 31_-0 to the first branched-marching main memory 31_-1, from the first branched-marching main memory 31_-1 to the first

marching cache memories

21_-11, 21_-12,........., 21_-1p in parallel, from first

marching cache memories

21_-11, 21_-12,........., 21_-1p to the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p in parallel, and from the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p to the first

arithmetic pipelines

117_-11, 117_-12, ........., 117_-1p in parallel. The resultant data of the processing in the first

arithmetic pipelines

117_-11, 117_-12,........., 117_-1p are sent out to the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p so that data are transferred bi-directionally between the first marching-

vector register files

22v_-11, 22v_-12,........., 22v_-1p and the first

arithmetic pipelines

117_-11, 117_-12, ........., 117_-1p. Furthermore, the data stored in the first marching-

vector register files

22v_-11, 22v_-12, ........., 22v_-1p are sent out to the first

marching cache memories

21_-11, 21_-12,........., 21_-1p so that data are transferred bi-directionally between the first

marching cache memories

21_-11, 21_-12, ........., 21_-1p and the first marching-

vector register files

22v_-11, 22v_-12, ........., 22v_-1p, and the data stored in the first

marching cache memories

21_-11, 21_-12, ........., 21_-1p are sent out to the first branched-marching main memory 31_-1 so that data are transferred bi-directionally between the first branched-marching main memory 31_-1 and the first

marching cache memories

21_-11, 21_-12, ........., 21_-1p. However, the FPGA controls the movement of instructions such that there is no flow along the opposite direction of the information to be processed in the first processing unit 12_-1.

The second processing unit 12_-2 encompasses a second branched-marching main memory 31_-2, a plurality of second

marching cache memories

21_-21, 21_-22,........., 21_-2p electrically connected respectively to the second branched-marching main memory 31_-2, a plurality of second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q electrically connected respectively to the second

marching cache memories

21_-21, 21_-22,........., 21_-2p, a plurality of second

arithmetic pipelines

117_-21, 117_-22,........., 117_-2p electrically connected respectively to the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q. Similar to the first processing unit 12_-1, each of the mother marching main memory 31_-0, the second branched-marching main memory 31_-2, the second

marching cache memories

21_-21, 21_-22,........., 21_-2p, and the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2p encompasses an array of memory units, input terminals of the array and output terminals of the array, configured to store information in each of memory units and to transfer synchronously with the clock signal, step by step, from a side of input terminals toward the output terminals. Because the operations of the mother marching main memory 31_-0, the second branched-marching main memory 31_-2, the second

marching cache memories

21_-21, 21_-22,........., 21_-2p, and the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2pare controlled by the FPGA, the information moves from the mother marching main memory 31_-0 to the second branched-marching main memory 31_-2, from the second branched-marching main memory 31_-2 to the second

marching cache memories

21_-21, 21_-22,........., 21_-2q in parallel, from second

marching cache memories

21_-21, 21_-22,........., 21_-2q to the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q in parallel, and from the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q to the second

arithmetic pipelines

117_-21, 117_-22, ........., 117_-2q in parallel. The resultant data of the processing in the second

arithmetic pipelines

117_-21, 117_-22,........., 117_-2q are sent out to the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q so that data are transferred bi-directionally between the second marching-

vector register files

22v_-21, 22v_-22,........., 22v_-2q and the second

arithmetic pipelines

117_-21, 117_-22, ........., 117_-2q. Furthermore, the data stored in the second marching-

vector register files

22v_-21, 22v_-22, ........., 22v_-2q are sent out to the second

marching cache memories

21_-21, 21_-22,........., 21_-2q so that data are transferred bi-directionally between the second

marching cache memories

21_-21, 21_-22, ........., 21_-2q and the second marching-

vector register files

22v_-21, 22v_-22, ........., 22v_-2q, and the data stored in the second

marching cache memories

21_-21, 21_-22, ........., 21_-2q are sent out to the second branched-marching main memory 31_-2 so that data are transferred bi-directionally between the second branched-marching main memory 31_-2 and the second

marching cache memories

21_-21, 21_-22, ........., 21_-2q. However, the FPGA controls the movement of instructions such that there is no flow along the opposite direction of the information to be processed in the second processing unit 12_-2.

For example, vector instructions generated from loops in a source program are transferred from the mother marching main memory 31_-0 to the first processing unit 12_-1, the second processing unit 12_-2, the third processing unit 12_-3, ..........in parallel, so that parallel processing of these vector instructions can be executed by arithmetic pipelines117_-11, 117_-12, ........., 117_-1p, 117_-21, 117_-22, ........., 117_-2q, .......... in each of the first processing unit 12_-1, the second processing unit 12_-2, the third processing unit 12_-3,........

Although the current FPGA-controlled HPC system requires a large amount of wiring resources, which generate time delay and stray capacitances between these wires, contributing the bottleneck, in the HPC system of the fourth embodiment illustrated in Fig. 63, because there are no buses such as data bus and address bus for any data exchange between the first marching-vector register files 22v_-11, 22v_-12,........., 22v_-1p and the first arithmetic pipelines 117_-11, 117_-12, ........., 117_-1p, between the first marching cache memories 21_-11, 21_-12, ........., 21_-1p and the first marching-vector register files 22v_-11, 22v_-12, ........., 22v_-1p, between the first branched-marching main memory 31_-1 and the first marching cache memories 21_-11, 21_-12, ........., 21_-1p, between the second marching-vector register files 22v_-21, 22v_-22,........., 22v_-2q and the second arithmetic pipelines 117_-21, 117_-22, ........., 117_-2q, between the second marching cache memories 21_-21, 21_-22, ........., 21_-2q and the second marching-vector register files 22v_-21, 22v_-22, ........., 22v_-2q, between the second branched-marching main memory 31_-2 and the second marching cache memories 21_-21, 21_-22, ........., 21_-2q, between the mother marching main memory 31_-0 and the first branched-marching main memory 31_-1, and between the mother marching main memory 31_-0 and the second branched-marching main memory 31_-2, the FPGA-controlled HPC system illustrated in Fig. 63 can achieve much higher processing speed and lower power consumption than the current FPGA-controlled HPC system. By increasing the number of

processing units

12_-1, 12_-2, 12_-3,........., the FPGA-controlled HPC system pertaining to the fourth embodiment can execute, for example, thousands of threads or more simultaneously at very high speed, enabling high computational throughput across large amounts of data.

--FIFTH EMBODIMENT--
As illustrated in Fig. 64, a computer system pertaining to a fifth embodiment of the present invention encompasses a processor 11, a stack of marching-

register files

22_-1, 22_-2, 22_-3, ........, implementing a three-dimensional marching-register file connected to the processor 11, a stack of marching-

cache memories

21_-1, 21_-2, 21_-3, ........, implementing a three-dimensional marching-cache memory connected to the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........), and a stack of marching

main memories

31_-1, 31_-2, 31_-3, ........., implementing a three-dimensional marching main memory connected to the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........). The processor 11 includes a control unit 111 having a clock generator 113 configured to generate a clock signal, an arithmetic logic unit (ALU) 112 configured to execute arithmetic and logic operations synchronized with the clock signal.

In the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........), a first marching-register file 22_-1 includes a first marching-instruction register file 22a_-1 connected to the control unit 111 and a first marching-data register file 22b_-1 connected to the ALU 112, a second marching-register file 22_-2 includes a second marching-instruction register file connected to the control unit 111 and a second marching-data register file connected to the ALU 112, a third marching-register file 22_-3 includes a third marching-instruction register file connected to the control unit 111 and a third marching-data register file connected to the ALU 112, and,......... In the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........), the first marching-cache memory 21_-1includes a first marching-instruction cache memory 21a_-1 and a first marching-data cache memory21b_-1, the second marching-cache memory 21_-2includes a second marching-instruction cache memory and a second marching-data cache memory, the third marching-cache memory 21_-3includes a third marching-instruction cache memory and a third marching-data cache memory, and.........

Although the illustration is omitted, very similar to the marching main memory 31 illustrated in Figs. 45-51, each of the marching main memories 31_-1, 31_-2, 31_-3, ........, has a two-dimensional array of memory units each having a unit of information, input terminals of the main memory array and output terminals of the main memory array, each of the marching main memories 31_-1, 31_-2, 31_-3, ........, stores the information in each of memory units and to transfer synchronously with the clock signal, step by step, toward the output terminals of the main memory array, so as to provide the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) with the stored information actively and sequentially, each of the marching-cache memories 21_-1, 21_-2, 21_-3, ........, has a two-dimensional array of cache memory units, cache input terminals of the marching-cache array configured to receive the stored information from the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, ........), and cache output terminals of the marching-cache array, each of the marching-cache memories 21_-1, 21_-2, 21_-3, ........, stores the information in each of cache memory units and to transfer, synchronously with the clock signal, step by step, the information to an adjacent cache memory unit, so as to provide actively and sequentially the stored information to the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........), and each of the marching-register files 22_-1, 22_-2, 22_-3, ........, has a two-dimensional array of register units each having a unit of information, input terminals of the register array configured to receive the stored information from the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........), and output terminals of the register array, each of the marching-register files 22_-1, 22_-2, 22_-3, ........, stores the information in each of register units and to transfer synchronously with the clock signal, step by step, toward the output terminals of the register array, so as to provide the processor 11 with the stored information actively and sequentially so that the processor 11 can execute the arithmetic and logic operations with the stored information.

Each of the marching

main memories

31_-1, 31_-2, 31_-3, ........, is implemented by the two-dimensional array of memory units delineated at a surface of a semiconductor chip, and a plurality of the semiconductor chips are stacked vertically as illustrated in 27A, sandwiching

heat dissipating plates

58m_-1, 58m_-2, 58m_-3, ......... between the plurality of the semiconductor chips so as to implement the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........). It is preferable that the

heat dissipating plates

58m_-1, 58m_-2, 58m_-3, ........., are made of materials having high thermal conductivity such as diamond. Similarly, each of the marching-

cache memories

21_-1, 21_-2, 21_-3, ........, is implemented by the two-dimensional array of memory units delineated at a surface of a semiconductor chip, and a plurality of the semiconductor chips are stacked vertically as illustrated in 27B, sandwiching

heat dissipating plates

58c_-1, 58c_-2, 58c_-3, ........., between the plurality of the semiconductor chips so as to implement the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........), and each of the marching-

register files

22_-1, 22_-2, 22_-3,........, is implemented by the two-dimensional array of memory units delineated at a surface of a semiconductor chip, and a plurality of the semiconductor chips are stacked vertically as illustrated in 27C, sandwiching

heat dissipating plates

58r_-1, 58r_-2, 58r_-3, ........., between the plurality of the semiconductor chips so as to implement the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........). It is preferable that the

heat dissipating plates

58c_-1, 58c_-2, 58c_-3, ........, 58r_-1, 58r_-2, 58r_-3, ........., are made of materials having high thermal conductivity such as diamond. Because there are no interconnects inside the surfaces of the semiconductor chips in the three-dimensional configuration illustrated in Figs. 65(a)-(c) and 66, it is easy to insert the

heat dissipating plates

58c_-1, 58c_-2, 58c_-3, ........., 58c_-1, 58c_-2, 58c_-3, ........, 58r_-1, 58r_-2, 58r_-3, ........., between the semiconductor chips, the configuration illustrated in Figs. 65(a)-(c) and 66 is expandable to stacking structures with any number of the semiconductor chips. In the conventional architecture, basically there is a limit of the number of stacked semiconductor chips in terms of thermal issues when the conventional semiconductor chips are directly stacked. In the computer system of the fifth embodiment, the sandwich structure illustrated in Figs. 65(a)-(c) and 66 is suitable for establishing the thermal flow from active computing semiconductor chips through the

heat dissipating plates

58c_-1, 58c_-2, 58c_-3, ........., 58c_-1, 58c_-2, 58c_-3, ........, 58r_-1, 58r_-2, 58r_-3, ........., to outside the system more effectively. Therefore, in the computer system of the fifth embodiment, these semiconductor chips can be stacked proportionally to the scale of the system, and as illustrated in Figs. 65(a)-(c) and 66, because a plurality of the semiconductor chips merging the marching

main memories

31_-1, 31_-2, 31_-3, ........, the marching-

cache memories

21_-1, 21_-2, 21_-3, ........, and the marching-

register files

22_-1, 22_-2, 22_-3,........, could easily be stacked to implement the three-dimensional configuration, a scalable computer systems can be easily organized, thereby keeping the temperature of the system cooler.

Although the illustration is omitted, the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) and the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) are electrically connected by a plurality of joint members, the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) and the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) are electrically connected by a plurality of joint members, and the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) and processor 11 are electrically connected by another plurality of joint members.
The resultant data of the processing in the ALU 112 are sent out to the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) through the joint members so that data are transferred bi-directionally between the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) and the ALU 112. Furthermore, the data stored in the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) are sent out to the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) through the joint members so that data are transferred bi-directionally between the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) and the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........). Furthermore, the data stored in the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) are sent out to the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) through the joint members so that data are transferred bi-directionally between the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) and the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........).

On the contrary, there is only one way of instruction-flow from the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) to the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........), from the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) to the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........), and from the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) to the control unit 111. For example, vector instructions generated from loops in a source program are transferred from the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) to the control unit 111 through the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) through the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) and the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) so that each of these vector instructions can be executed by arithmetic pipelines in the control unit 111.
In the computer system of the fifth embodiment illustrated in Fig. 64, there are no buses such as the data bus and address bus in any data exchange between the three-dimensional marching main memory (31_-1, 31_-2, 31_-3, .........) and the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........), between the three-dimensional marching-cache (21_-1, 21_-2, 21_-3, ........) and the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........), and between the three-dimensional marching-register file (22_-1, 22_-2, 22_-3, ........) and the processor 11, while the wires or the buses implement the bottleneck in the conventional computer system. As there are no global wires, which generate time delay and stray capacitances between these wires, the computer system of the fifth embodiment can achieve much higher processing speed and lower power consumption than the conventional computer system, keeping the temperature of the computer system at lower temperature than the conventional computer system so as to establish "a cool computer", by employing the

heat dissipating plates

58c_-1, 58c_-2, 58c_-3, ........., 58c_-1, 58c_-2, 58c_-3, ........, 58r_-1, 58r_-2, 58r_-3, ........., which are made of materials having high thermal conductivity such as diamond and disposed between the semiconductor chips. The cool computer pertaining to the fifth embodiment is different from existing computers because the cool computer is purposely architected and designed with an average of 30% less energy consumption and 10000% less size to obtain 100 times higher speed, for example.

Since other functions, configurations, way of operation of the computer system pertaining to the fifth embodiment are substantially similar to the functions, configurations, way of operation already explained in the first to third embodiments, overlapping or redundant description may be omitted.

(MISCELLANEOUS THREE-DIMENSIONAL CONFIGURATIONS)
The three-dimensional configurations illustrated in Figs. 64, 65(a), 65(b) and 65(c) are mere examples, and there are various ways and combinations how to implement three-dimensional configurations so as to facilitate the organization of a scalable computer system.

For example, as illustrated in Fig. 66, a first chip (top chip) merging a plurality of arithmetic pipelines 117 and a plurality of marching-register files 22, a second chip (middle chip) merging a marching-cache memory 21 and a third chip (bottom chip) merging a marching main memory 31 can be stacked vertically. Each of the arithmetic pipelines 117 may include a vector-processing unit, and each of the marching-register files 22 may include marching-vector registers. Between the first and second chips, a plurality of joint members 55a are inserted, and between the second and third chips, a plurality of joint members 55b are inserted. For example, each of

joint members

55a and 55b may be implemented by an electrical conductive bump such as a solder ball, a gold (Au) bump, a silver (Ag) bump, a copper (Cu) bump, a nickel-gold (Ni-Au) alloy bump or a nickel-gold-indium (Ni-Au-In) alloy bump. Although the illustration is omitted, heat-dissipating plates can be inserted between the first and second chips and between the second and third chips so as to achieve "cool chips", similar to the configuration illustrated in Figs. 65(a)-(c) and 66.

Alternatively, as illustrated in Figs. 67 and 68, a first three-dimensional (3D)-stack embracing a first top chip, a first middle chip and first bottom chip and a second 3D-stack embracing a second top chip, a second middle chip and second bottom chip may be disposed two dimensionally on a same substrate or a same circuit board so as to implement a parallel computing with multiple processors, in which the first 3D-stack and the second 3D-stack are connected by

bridges

59a and 59b.

In the first 3D-stack, a first top chip merging a plurality of first arithmetic pipelines 117_-1 and a plurality of first marching-register files 22_-1, a first middle chip merging a first marching-cache memory 21_-1 and a first bottom chip merging a first marching main memory 31_-1 are 3D-stacked vertically. Each of the first arithmetic pipelines 117_-1 may include a vector-processing unit, and each of the first marching-cache files 22_-1 may include marching-vector registers. Between the first top and first middle chips, a plurality of joint members 55a_-1 are inserted, and between the first middle and first bottom chips, a plurality of joint members 55b_-1 are inserted. For example, each of

joint members

55a_-1 and 55b_-1 may be implemented by an electrical conductive bump such as a solder ball, a gold (Au) bump, a silver (Ag) bump, a copper (Cu) bump, a nickel-gold (Ni-Au) alloy bump or a nickel-gold-indium (Ni-Au-In) alloy bump. Similarly, in the second 3D-stack, a second top chip merging a plurality of second arithmetic pipelines 117_-2 and a plurality of second marching-register files 22_-2, a second middle chip merging a second marching-cache memory 21_-2 and a second bottom chip merging a second marching main memory 31_-2 are 3D-stacked vertically. Each of the second arithmetic pipelines 117_-2 may include a vector-processing unit, and each of the second marching-cache files 22_-2 may include marching-vector registers. Between the second top and second middle chips, a plurality of joint members 55a_-2 are inserted, and between the second middle and second bottom chips, a plurality of joint members 55b_-2 are inserted. For example, each of

joint members

55a_-2 and 55b_-2 may be implemented by an electrical conductive bump such as a solder ball, a gold (Au) bump, a silver (Ag) bump, a copper (Cu) bump, a nickel-gold (Ni-Au) alloy bump or a nickel-gold-indium (Ni-Au-In) alloy bump. Although the illustration is omitted, heat-dissipating plates can be inserted between the first top and first middle chips, between the first middle and first bottom chips, between the second top and second middle chips and between the second middle and second bottom chips similar to the configuration illustrated in Figs. 65(a)-(c) and 66 so as to achieve "cool chips".

Similar to the computer system of the fourth embodiment, a field programmable gate array (FPGA) may switch-control the operations of the first and second 3D-stacks, by traveling a thread or chaining of vector processing on the first arithmetic pipelines 117_-1 and the second arithmetic pipelines 117_-2, implementing a HPC system, which can be used for GPU-based general-purpose computing.

Still alternatively, as illustrated in Fig. 69, a first chip (top chip) merging a plurality of arithmetic pipelines 117, a second chip merging and a plurality of marching-register files 22, a third chip merging a marching-cache memory 21, a fourth chip merging a first marching main memory 31_-1, a fifth chip merging a marching main memory 31_-2 and a sixth chip (bottom chip) merging a third marching main memory 31_-3 can be stacked vertically. Each of the arithmetic pipelines 117 may include a vector-processing unit, and each of the marching-register files 22 may include marching-vector registers so that vector instructions generated from loops in a source program can be executed in the vector-processing unit. A first heat dissipating plate 58_-1 is inserted between the first and second chips, a second heat dissipating plate 58_-2 is between the second and third chips, a third heat dissipating plate 58_-1 is between the third and fourth chips, a fourth heat dissipating plate 58_-4 is between the fourth and fifth chips, and a fifth heat dissipating plate 58_-5 is between the fifth and sixth chips so as to achieve "cool chips". Because there are no interconnects inside the surfaces of these cool chips in the three-dimensional configuration illustrated in Fig. 69, it is easy to insert the

heat dissipating plates

58_-1, 58_-2, 58_-3, 58_-4, 58_-5 such as diamond chips between these six chips alternately.

The cool-chip-configuration illustrated in Fig. 69 is not limited to a case of six chips, but expandable to three-dimensional stacking structures with any number of chips, because the sandwich structure illustrated in Fig. 69 is suitable for establishing the thermal flow from active computing chips through the

heat dissipating plates

58_-1, 58_-2, 58_-3, 58_-4, 58_-5 to outside of the cool computer system more effectively. Therefore, the number of cool chips in the computer system of the fifth embodiment can be increased in proportion to the scale of the computer system.

Figs. 70-72 show various examples of the three-dimensional (3D) stack, implementing a part of fundamental cores of the computer systems according to the fifth embodiment of the present invention, each of the 3D-stacks includes cooling technology with heat dissipating plate 58 such as diamond plate inserted between the

semiconductor memory chips

3a and 3b, in which at least one of the marching memory classified in the marching memory family is merged, the term of "the marching memory family" includes the marching-instruction register file 22a and the marching-data register file 22b connected to the ALU 112 explained in the second embodiment, and the marching-instruction cache memory 21a and the marching-data cache memory 21b explained in the third embodiment, in addition to the marching main memory 31 explained in the first embodiment of the present invention.

That is, as illustrated in Fig. 70, a 3D-stack, implementing a part of the fundamental core of the computer system pertaining to the fifth embodiment of the present invention, embraces a first semiconductor memory chip 3a merging at least one of the marching memory in the marching memory family, a heat dissipating plate 58 disposed under the first semiconductor memory chip 3a, a second semiconductor memory chip 3b disposed under the heat dissipating plate 58, which merges at least one of the marching memory in the marching memory family, and a processor 11 disposed at a side of the heat dissipating plate 58. Here, in Fig. 70, because the location of the processor 11 is illustrated as one of the examples, the processor 11 can be disposed at any required or appropriate site in the configuration of the 3D-stack or external of the 3D-stack, depending on the design choice of the 3D-stack. For example, the processor 11 can be allocated at the same horizontal level of the first semiconductor memory chip 3a or at the level of the second semiconductor memory chip 3b. The marching memory merged on the first semiconductor memory chip 3a and the marching memory merged on the second semiconductor memory chip 3b stores program instruction, respectively. In the 3D configuration illustrated in Fig. 70, in which the first semiconductor memory chip 3a, the heat dissipating plate 58 and the second semiconductor memory chip 3b are stacked vertically, a first control path is provided between the first semiconductor memory chip 3a and the processor 11, and a second control path is provided between the second semiconductor memory chip 3b and the processor 11 so as to facilitate the execution of the control processing with the processor 11. A further data-path may be provided between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b so as to facilitate direct communication of the program instruction between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b.

And, as illustrated in Fig. 71, another 3D-stack, implementing a part of the fundamental core of the computer system pertaining to the fifth embodiment of the present invention, embraces a first semiconductor memory chip 3a merging at least one of the marching memory in the marching memory family, a heat dissipating plate 58 disposed under the first semiconductor memory chip 3a, a second semiconductor memory chip 3b disposed under the heat dissipating plate 58, which merges at least one of the marching memory in the marching memory family, and a ALU 112 disposed at a side of the heat dissipating plate 58. The location of the ALU 112 is not limited to the site illustrated in Fig. 71, and the ALU 112 can be disposed at any required or appropriate site in the configuration of the 3D-stack or external of the 3D-stack, such as a site allocated at the same horizontal level of the first semiconductor memory chip 3a or at the level of the second semiconductor memory chip 3b, depending on the design choice of the 3D-stack. The marching memory merged on the first semiconductor memory chip 3a and the marching memory merged on the second semiconductor memory chip 3b read/write scalar data, respectively. In the 3D configuration illustrated in Fig. 71, in which the first semiconductor memory chip 3a, the heat dissipating plate 58 and the second semiconductor memory chip 3b are stacked vertically, a first data-path is provided between the first semiconductor memory chip 3a and the ALU 112, and a second data-path is provided between the second semiconductor memory chip 3b and the ALU 112 so as to facilitate the execution of the scalar data processing with the ALU 112. A further data-path may be provided between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b so as to facilitate direct communication of the scalar data between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b.

Further, as illustrated in Fig. 72, a still another 3D-stack, implementing a part of the fundamental core of the computer system pertaining to the fifth embodiment of the present invention, embraces a first semiconductor memory chip 3a merging at least one of the marching memory in the marching memory family, a heat dissipating plate 58 disposed under the first semiconductor memory chip 3a, a second semiconductor memory chip 3b disposed under the heat dissipating plate 58, which merges at least one of the marching memory in the marching memory family, and an arithmetic pipelines 117 disposed at a side of the heat dissipating plate 58. Similar to the topologies illustrated in Figs. 62 and 63, the location of the arithmetic pipelines 117 is not limited to the site illustrated in Fig. 72, and the arithmetic pipelines 117 can be disposed at any required or appropriate site. The marching memory merged on the first semiconductor memory chip 3a and the marching memory merged on the second semiconductor memory chip 3b read/write vector/streaming data, respectively. In the 3D configuration illustrated in Fig. 72, in which the first semiconductor memory chip 3a, the heat dissipating plate 58 and the second semiconductor memory chip 3b are stacked vertically, a first data-path is provided between the first semiconductor memory chip 3a and the arithmetic pipelines 117, and a second data-path is provided between the second semiconductor memory chip 3b and the arithmetic pipelines 117 so as to facilitate the execution of the vector/streaming data processing with the arithmetic pipelines 117. A further data-path may be provided between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b so as to facilitate direct communication of the vector/streaming data between the first semiconductor memory chip 3a and the second semiconductor memory chip 3b.

As illustrated in Fig. 73, the 3D hybrid computer system according to the fifth embodiment encompasses a first left chip (top left chip) 3p_-1 merging at least one of the marching memory in the marching memory family, a second left chip 3p_-2 merging at least one of the marching memory in the marching memory family, a third left chip 3p_-3 merging at least one of the marching memory in the marching memory family, a fourth left chip 3p_-4 merging at least one of the marching memory in the marching memory family, a fifth left chip 3p_-5 merging at least one of the marching memory in the marching memory family and a sixth left chip (bottom left chip) 3p_-6merging at least one of the marching memory in the marching memory family, which are stacked vertically. A first left heat dissipating plate 58a_-1 is inserted between the first left chip 3p_-1 and second left chip 3p_-2, a second left heat dissipating plate 58a_-2 is inserted between the second left chip 3p_-2 and third left chip 3p_-3, a third left heat dissipating plate 58a_-1 is inserted between the third left chip 3p_-3 and fourth left chip 3p_-4, a fourth left heat dissipating plate 58a_-4 is inserted between the fourth left chip 3p_-4 and fifth left chip 3p_-5, and a fifth left heat dissipating plate 58a_-5 is inserted between the fifth left chip 3p_-5 and sixth left chip 3p_-6 so as to achieve "cool left chips".
And, a first right chip (top right chip) 3q_-1 merging at least one of the marching memory in the marching memory family, a second right chip 3q_-2 merging at least one of the marching memory in the marching memory family, a third right chip 3q_-3 merging at least one of the marching memory in the marching memory family, a fourth right chip 3q_-4 merging at least one of the marching memory in the marching memory family, a fifth right chip 3q_-5 merging at least one of the marching memory in the marching memory family and a sixth right chip (bottom right chip) 3q_-6merging at least one of the marching memory in the marching memory family are stacked vertically. A first right heat dissipating plate 58b_-1 is inserted between the first right chip 3q_-1 and second right chip 3q_-2, a second right heat dissipating plate 58b_-2 is inserted between the second right chip 3q_-2 and third right chip 3q_-3, a third right heat dissipating plate 58b_-1 is inserted between the third right chip 3q_-3 and fourth right chip 3q_-4, a fourth right heat dissipating plate 58b_-4 is inserted between the fourth right chip 3q_-4 and fifth right chip 3q_-5, and a fifth right heat dissipating plate 58b_-5 is inserted between the fifth right chip 3q_-5 and sixth right chip 3q_-6 so as to achieve "cool right chips".

A first processing unit 11a is provided between the first left heat dissipating plate 58a_-1 and the first right heat dissipating plate 58b_-1, a second processing unit 11b is provided between the third left heat dissipating plate 58a_-3 and the third right heat dissipating plate 58b_-3, and a third processing unit 11c is provided between the fifth left heat dissipating plate 58a_-5 and the fifth right heat dissipating plate 58b_-5, and pipelined ALUs are respectively included in the processing units11a, 11b, 11c.

The scalar data-path and control path are established between the first left chip 3p_-1 and second left chip 3p_-2, the scalar data-path and control path are established between the second left chip 3p_-2 and third left chip 3p_-3, the scalar data-path and control path are established between the third left chip 3p_-3 and fourth left chip 3p_-4, the scalar data-path and control path are established between the fourth left chip 3p_-4 and fifth left chip 3p_-5, and the scalar data-path and control path are established between the fifth left chip 3p_-5 and sixth left chip 3p_-6,the scalar data-path and control path are established between the first right chip 3q_-1 and second right chip 3q_-2, the scalar data-path and control path are established between the second right chip 3q_-2 and third right chip 3q_-3, the scalar data-path and control path are established between the third right chip 3q_-3 and fourth right chip 3q_-4, the scalar data-path and control path are established between the fourth right chip 3q_-4 and fifth right chip 3q_-5, and the scalar data-path and control path are established between the fifth right chip 3q_-5 and sixth right chip 3q_-6. The 3D computer system illustrated in Fig. 73 can execute not only scalar data but also vector/streaming data through the combination of scalar data-path and control path for the computer system.

Because there are no interconnects inside the surfaces of these cool chips in the 3D configuration illustrated in Fig. 73, it is easy to insert the

heat dissipating plates

58a_-1, 58a_-2, 58a_-3, 58a_-4, 58a_-5 such as diamond left chips between these six left chips alternately, and to insert the

heat dissipating plates

58b_-1, 58b_-2, 58b_-3, 58b_-4, 58b_-5 such as diamond right chips between these six right chips alternately.

--OTHER EMBODIMENTS--
Various modifications will become possible for those skilled in the art after receiving the teaching of the present disclosure without departing from the scope thereof.

In Figs. 4, 5, 6, 8, 11, 13, 16-20, 22, 25 and 32, although nMOS transistors are assigned respectively as the transfer-transistors and the reset-transistors in the transistor-level representations of the bit-level cells, because the illustration in Figs. 4, 5, 6, 8, 11, 13, 16-20, 22, 25 and 32 are mere schematic examples, pMOS transistors can be used as the transfer-transistors and the reset-transistors, if the opposite polarity of the clock signal is employed. Furthermore, MIS transistors, or insulated-gate transistors having gate-insulation films made of silicon nitride film, ONO film, SrO film, Al₂O₃ film, MgO film, Y₂O₃ film, HfO₂ film, ZrO₂ film, Ta₂O₅ film, Bi₂O₃ film, HfAlO film, and others can be used for the transfer-transistors and the reset-transistors.

There are several different forms of parallel computing such as bit-level, instruction level, data, and task parallelism, and as well known as "Flynn's taxonomy", programs and computers are classified as to whether they were operating using a single set or multiple sets of instructions, whether or not those instructions were using a single or multiple sets of data.

For example, as illustrated in Fig. 74, a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments can implement a bit-level parallel processing of scalar/vector data in a multiple-instruction-single-data (MISD) architecture, by which many independent instruction streams provided vertically to a first processor 11_-1, a second processor 11_-2, a third processor 11_-3, a fourth processor 11_-4, .........., in parallel operate on a single horizontal stream of data at a time with a systolic array of

processors

11_-1, 11_-2, 11_-3, 11_-4 .

Alternatively, as illustrated in Fig. 75,arithmetic-level parallelism can be established by a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments, with a single-instruction-multiple-data (SIMD) architecture, by which a single instruction stream is provided to a first processor 11_-1, a second processor 11_-2, a third processor 11_-3, and a fourth processor 11_-4, so that the single instruction stream can operate on multiple vertical streams of data at a time with the array of

processors

11_-1, 11_-2, 11_-3, 11_-4 .

Still alternatively, as illustrated in Fig. 76, a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments, can implement a typical chaining in vector processing with a first processor 11_-1, a second processor 11_-2, a third processor 11_-3, and a fourth processor 11_-4 to which a first instruction I₁, a second instruction I₂, a third instruction I₃, and a fourth instruction I₄ are provided respectively.

Furthermore, as illustrated in Fig. 77, a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments, can implement a parallel processing of a single horizontal stream of scalar/vector data in a MISD architecture with a first processor 11_-1, a second processor 11_-2, a third processor 11_-3, and a fourth processor 11_-4.

Furthermore, as illustrated in Fig. 78, a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments, can implement a parallel processing of a single horizontal stream of scalar/vector data in a MISD architecture with a first processor 11_-1configured execute multiplication, a second processor 11_-2 configured execute addition, a third processor 11_-3 configured execute multiplication, and a fourth processor 11_-4 configured execute addition.

Furthermore, as to process-level parallelism, a single-thread-stream and single-data-stream architecture, a single-thread-stream and multiple-data-streams architecture, a multiple-thread-streams and single-data-stream architecture, and a multiple-thread-streams and multiple-data-streams architecture can be achieved with a marching memory, which may include the marching-register file, the marching-cache memory, and the marching main memory already discussed in the first to fifth embodiments.

Referring to Fig. 41, we have compared the speed/capability of the worst case of the existing memory for scalar data or program instructions with that of the marching main memory 31, an the hatched portion of Fig. 41(b) has illustrated schematically the speed/capability of the marching main memory 31, implemented by one hundred of memory units U₁, U₂, U₃,........., U₁₀₀, and compared with the speed/capability of the worst case of the existing memory shown in Fig. 41(a). In the worst case, we have discussed that we can read out 99 memory units of the marching main memory 31, but they are not available due to a scalar program's requirement. However, with "a complex marching memory" scheme illustrated in Fig. 79(b), we can improve the speed/capability of the marching memory for scalar data or program instructions, in which a plurality of marching memory blocks MM₁₁, MM₁₂, MM₁₃, ........, MM₁₆;MM₂₁, MM₂₂, MM₂₃, ........, MM₂₆;MM₃₁, MM₃₂, MM₃₃, ........, MM₃₆;..................;MM₅₁, MM₅₂, MM₅₃, ........, MM₅₆ are deployed two dimensionally and merged on a single semiconductor chip 66, and a specified marching memory block MM_ij (i=1 to 5; j=1 to 6) can be randomly accessed from the plurality of marching memory blocks MM₁₁, MM₁₂, MM₁₃, ........, MM₁₆;MM₂₁, MM₂₂, MM₂₃, ........, MM₂₆;MM₃₁, MM₃₂, MM₃₃, ........, MM₃₆;..................;MM₅₁, MM₅₂, MM₅₃, ........, MM₅₆, similar to the random-access methodology employed in a dynamic random access memory (DRAM) architecture.

As illustrated in Fig. 79(a), in a conventional DRAM, a memory array area 661, peripheral circuitry for a row decoder 662, peripheral circuitry for sense amplifiers 663, and peripheral circuitry for a column decoder 664 are merged on a single semiconductor chip 66. A plurality of memory cells are arranged in an array of rows and columns in the memory array area 661 so that each row of memory cells share a common 'word' line, while each column of cells share a common 'bit' line, and the location of a memory cell in the array is determined as the intersection of its 'word' and 'bit' lines. During a 'write' operation, the data to be written ('1' or '0') is provided at the 'bit' line from the column decoder 664, while the 'word line' is asserted from the row decoder 662, so as to turn on the access transistor of the memory cell and allows the capacitor to charge up or discharge, depending on the state of the bit line. During a 'read' operation, the 'word' line is also asserted from the row decoder 662, which turns on the access transistor. The enabled transistor allows the voltage on the capacitor to be read by a sense amplifier 663 through the 'bit' line. The sense amplifier 663 can determine whether a '1' or '0' is stored in the memory cell by comparing the sensed capacitor voltage against a threshold.

Although 6*5=30 marching memory blocks MM₁₁, MM₁₂, MM₁₃, ........, MM₁₆;MM₂₁, MM₂₂, MM₂₃, ........, MM₂₆;MM₃₁, MM₃₂, MM₃₃, ........, MM₃₆;..................;MM₅₁, MM₅₂, MM₅₃, ........, MM₅₆ are deployed on the semiconductor chip 66 for avoiding cluttering up the drawings, the illustration is schematic, and actually one thousand marching memory blocks MM_ij (i=1 to s; j=1 to t; and s*t =1000) with 256kbits capacity can be deployed on the same semiconductor chip 66, if unidirectional marching memories are arrayed, and if 512 Mbits DRAM chip technology is assumed as the manufacturing technology of the complex marching memory scheme illustrated in Fig. 79(b). That is, as an area for monolithically integrating each of the marching memory blocks MM_ij having 256kbits capacity on a semiconductor chip 66, an equivalent area for 512kbits DRAM block is required, because, as illustrated in Figs. 4-6, each of unidirectional marching memory blocks is implemented by a bit-level cell consisting of two transistors and one capacitor, while the DRAM memory cell consists of only a single transistor that is paired with a capacitor. Alternatively, as to an array of bidirectional marching memories, one thousand marching memory blocks MM_ij with 128kbits capacity can be deployed on the same semiconductor chip 66 for the 512 Mbits DRAM chip. That is, as an area for monolithically integrating each of the marching memory blocks MM_ij having 128kbits capacity, an equivalent area for the 512kbits DRAM block is required, because, as illustrated in Fig. 32, a bidirectional marching memory block is implemented by a bit-level cell consisting of four transistors and two capacitors, while the DRAM memory cell consists of only a single transistor and a single capacitor. If one Gbit DRAM chip technology is assumed, one thousand bidirectional marching memory blocks MM_ij with 256kbits capacity can be deployed on the same DRAM chip 66 so as to implement a 256 Mbits marching memory chip.

Therefore, one thousand of marching memory blocks MM_ij, or one thousand of marching memory cores can be monolithically integrated on the semiconductor chip 66, as illustrated in Fig. 79(b). A single marching memory block MM_ij, or "a single marching memory core" may encompass, for example, one thousand of marching memory columns, or one thousand of marching memory units U_k (k=1 to 1000), which have 1000*32 byte-based addresses, where one memory unit U_k has 256 bit-level cells. That is, with a complex marching memory chip having one thousand of marching memory blocks MM_ij, one thousand of marching memory units U_k (k=1 to 1000) of 32 bytes (or 256 bits) are allowable to access within one cycle of the conventional DRAM access.

Figs. 80 (a) and (b) illustrate an example of a single 256kbits marching memory blocks MM_ij, which has one thousand of marching memory units U_k (k=1 to n; n=1000) of 32 bytes (or 256 bits). In the complex marching memory schemes, as illustrated in Fig. 80 (b), position indexes T_k (k=1 to 1000) or position tags are labeled, respectively, on each of the marching memory units U_k as the token of each of the columns U_k that means the first address of the column bytes. In Fig. 80 (b), the clock period (the clock cycle time) TAU(Greek-letter)_clock, illustrated in Fig. 7C, is recited as "the marching memory's memory cycle t_M".

In the light of above discussions stated in the first to fifth embodiments, because we can use so large speed difference between the conventional DRAM and the marching memory, as illustrated in Fig. 80 (c), with the conventional DRAM's memory cycle t_C for writing in or reading out the content of the conventional DRAM's one memory element, we can estimate as:

t_C = 1000t_M ............(1) .

Therefore, with the complex marching memory scheme illustrated in Fig. 79(b), we can improve the speed/capability of the marching memory for scalar data or program instructions, by which a specified marching memory block MM_ij (i=1 to s; j=1 to t; and s*t =1000) can be randomly accessed from one thousand of marching memory blocks, similar to the random-access methodology employed in the DRAM architecture.

Although the illustration is omitted in Fig. 79(b), the plurality of 256 kbits marching memory blocks MM_ij may be arranged in the two dimensional matrix form on the semiconductor chip 66 so that each horizontal array of the marching memory blocks MM_ij share a common horizontal-core line, while each vertical array of marching memory blocks MM_ij share a common vertical-core line, and a location of a specified marching memory block MM_ij in the two dimensional matrix is accessed as the intersection of its horizontal-core line and vertical-core line, with double-level hierarchy. In the double-level hierarchy, every column of a subject marching memory block MM_ij is accessed with an address at the lower level, and every marching memory block MM_ij are directly accessed with its own address for each marching memory block MM_ij at the higher level.

Alternatively, a virtual storage mechanism can be used for the access methodology of the complex marching memory. In the virtual storage mechanism, the marching memory blocks MM_ij (i=1 to s; j=1 to t), or the marching memory cores to be used are scheduled just like pages in a virtual memory. The scheduling is decided at compilation run if any. For example, in the multi-level caches architecture, the multi-level caches generally operate by checking the smallest Level1 (L1) cache first, and if the L1 cache hits, the processor proceeds at high speed. If the smaller L1 cache misses, the next larger cache (L2) is checked, and so on, before external memory is checked. For the access methodology of the complex marching memory, the L2 cache-like memory can support the virtual indexing mechanism, because the size of L2 cache corresponds to the size of the complex marching memory, and the size of a marching memory block MM_ij corresponds to the size of smallest L1 cache.

Then, because the achievement of the complex marching memory encompassing one thousand of marching memory blocks, or one thousand of cores is relatively easy as stated above, and in the complex marching memory, the access of any column is basically available at the CPU's clock rate, even at the worst case, the speed of the complex marching memory keeps the speed of the conventional DRAM.

Furthermore, a plurality of complex marching memory chips, or a plurality of macro complex marching memory blocks MMM₁, MMM₂, ........, MMM_k, can be mounted on a first circuit board having external-connection pins P₁, P₂, ........, P_s-1, P_s ("s" may be any integer determined by unit of byte, or word size) so as to implement a multichip module of the complex marching memory, or "a complex marching memory module" as illustrated in Fig. 81, although the illustration of the circuit board is omitted. In the hybrid assembly of macro complex marching memory blocks MMM₁, MMM₂, ........, MMM_k, the first macro complex marching memory block MMM₁ may monolithically integrate one thousand of marching memory blocks MM₁₁₁, MM₁₂₁, MM₁₃₁, ........, MM_1(t-1)1, MM_1t1; MM₂₁₁,........,; MM_(s-1)11..................; MM_s11, MM_s21, , ........, MM_s(t-1)1, MM_st1 on a first semiconductor chip, the second macro complex marching memory block MMM₂ may monolithically integrate one thousand of marching memory blocks MM₁₁₂, MM₁₂₂, MM₁₃₂, ........, MM_1(t-1)2, MM_1t2; MM₂₁₂,........,; MM_(s-1)12..................; MM_s12, MM_s22, , ........, MM_s(t-1)2, MM_st2 on a second semiconductor chip, ............., and the k-th macro complex marching memory block MMM_k may monolithically integrate one thousand of marching memory blocks MM_11k, MM_12k, MM_13k, ........, MM_1(t-1)k, MM_1tk; MM_21k,........,; MM_(s-1)1k..................; MM_s1k, MM_s2k, , ........, MM_s(t-1)k, MM_stk on a k-th semiconductor chip, for example. And the first complex marching memory module hybridly assembling the macro complex marching memory blocks MMM₁, MMM₂, ........, MMM_k can be connected to a second complex marching memory module hybridly assembling the macro complex marching memory block MMM_k+1 and others on a second circuit board through the external-connection pins P₁, P₂, ........, P_s-1, P_s. Here, the macro complex marching memory block MMM_k+1 may monolithically integrate one thousand of marching memory blocks MM_11(k+1), MM_12(k+1), MM_13(k+1), ........, MM_1(t-1)(k+1), MM_1t(k+1); MM_21(k+1),........,; MM_(s-1)1(k+1)..................; MM_s1(k+1), MM_s2(k+1), , ........, MM_s(t-1)(k+1), MM_st(k+1) on a semiconductor chip, for example. In addition, if we implement dual lines of the hybrid assembly of macro complex marching memory blocks, we can establish a dual in-line module of complex marching memory.

In the configuration of the complex marching memory modules illustrated in Fig. 81, by using triple-level hierarchy, every column of a subject marching memory block MM_iju (u=1 to k; "k" is any integer greater than or equal to two) is accessed with an address at the lowest level, every marching memory block MM_iju are accessed with its own address for each marching memory block MM_iju at the middle level, and every macro marching memory block MMM_u (u=1 to k) may be directly accessed with its own address at the highest level, which facilitate accessing to a remote column of the marching memory for scalar data or program instructions.

Alternatively, very similar to DRAM rank architecture encompassing a set of DRAM chips that operate in lockstep fashion to command in a memory, in which the DRAM chips inside the same rank are accessed simultaneously, the plurality of macro complex marching memory blocks MMM₁, MMM₂, ........, MMM_k, can be random accessed simultaneously, and with the above-mentioned double-level hierarchy methodology, every column of a subject marching memory block MM_iju (u=1 to k) is accessed with an address at the lower level, and every marching memory block MM_iju are directly accessed with its own address for each marching memory block MM_iju at the higher level.

Still alternatively, a virtual storage mechanism can be used for the access methodology of the complex marching memory, in which the marching memory cores to be used are scheduled just like pages in the virtual memory. The scheduling can be decided at compilation run if any.

Because the data transfer between the marching main memory 31 and the processor 11 is achieved at a very high speed, the cash memory employed in the conventional computer system is not required, and the cash memory can be omitted. However, similar to the organization illustrated in Fig. 56, a marching-data cache memory 21b implemented by the complex marching memory scheme can by used with more smaller size of marching memory blocks, or more smaller size of marching memory cores. For example, a plurality of marching memory cores with 1 kbits, 512 bits, or 256 bits capacity can be deployed on a semiconductor chip so as to implement the marching-data cache memory 21b, while a plurality of marching memory cores MM_ij (i=1 to s; j=1 to t; and s*t =1000) with 256 kbits capacity are deployed on the semiconductor chip 66 so as to implement marching main memory 31. And, for example, with the virtual storage mechanism, each of the marching memory cores can be randomly accessed.

Alternatively, one-dimensional array of marching memory blocks, or marching memory cores, being deployed vertically on a semiconductor chip, can implement a marching cache memory. Here, each of the marching memory cores includes a single horizontal array of memory units, and the number of memory units deployed horizontally is smaller than the number of memory units employed in the marching memory cores for the marching main memory 31. And, for example, with the virtual storage mechanism, each of the marching memory cores can be randomly accessed.

Furthermore, a plurality of marching memory blocks, or a plurality of marching memory cores can be deployed vertically on a semiconductor chip, each of the marching memory blocks consist of a single memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size so as to implement a marching register file by the complex marching memory scheme.

In the ultimate case of scaling the marching memory core, it can be considered that a plurality of marching memory cores with minimized size, or one bit capacity can be deployed on a semiconductor chip by the complex marching memory scheme, which may correspond to the structure of conventional SRAM. Therefore, marching-data register file 22b implemented by one-bit marching memory cores can be connected to the ALU 112, similar to the organizations illustrated in Figs. 55 and 56. Then, very similar to the operation of SRAM, each of the one-bit marching memory cores can be randomly accessed.

Thus, the present invention of course includes various embodiments and modifications and the like, which are not detailed above. Therefore, the scope of the present invention will be defined in the following claims.

The instant invention can be applied to industrial fields of various computer systems, which require higher speed and lower power consumption.

Claims

A marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a transfer-transistor having a first main-electrode connected to a clock signal supply line through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of the memory units, through a second delay element;
a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and
a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor,
wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of the memory units.
The marching memory of claim 1, wherein in each of the bit-level cells, when a clock signal is applied to the control-electrode of the reset-transistor, the reset-transistor discharges the signal charge already stored in the capacitor.
The marching memory of claim 1, wherein in each of the bit-level cells, after the signal charge stored in the capacitor has been discharged, the transfer-transistor becomes active delayed by a first delay time determined by the first delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the transfer-transistor, the transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second delay time determined by the second delay element to the capacitor.
The marching memory of claim 3, wherein the first delay time is a quarter of clock period of the clock signal, and the second delay time is a half of the clock period.
The marching memory of claim 1, wherein in the transfer-transistor, the control-electrode controls a current flowing between the first main-electrode and the second main-electrode electro-statically.
The marching memory of claim 1, wherein in the reset-transistor, the control-electrode controls a current flowing between the first main-electrode and the second main-electrode electro-statically.
The marching memory of claim 1, wherein the transfer-transistor and the reset-transistor are made of an insulated-gate transistor, respectively, including a MOS transistor, a MIS transistor and a high electron mobility transistor.
The marching memory of claim 7, wherein the transfer-transistor and the reset-transistor are made of an nMOS transistor, respectively, and the clock signal of positively high-level is applied to the control electrode of the nMOS transistor so as to achieve a conductive state.
The marching memory of claim 7, wherein the transfer-transistor and the reset-transistor are made of a pMOS transistor, respectively, and the clock signal of negatively high-level is applied to the control electrode of the nMOS transistor so as to achieve a conductive state.
A bidirectional-marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential;
a forward capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor; and
a backward capacitorconfigured to store the information of the bit-level cell, connected in parallel with the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the forward capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the backward capacitor to the first neighboring bit-level cell.
The bidirectional-marching memory of claim 10, wherein in each of the bit-level cells, when a first clock signal from a first clock signal supply line is applied to the control-electrode of the forward reset-transistor, the forward reset-transistor discharges the signal charge already stored in the forward capacitor, and a second clock signal from a second clock signal supply line is applied to the control-electrode of the backward reset-transistor, the backward reset-transistor discharges the signal charge already stored in the backward capacitor.
The bidirectional-marching memory of claim 10, wherein in each of the bit-level cells, after the signal charge stored in the forward capacitor has been discharged, the forward transfer-transistor becomes active delayed by a first forward delay time determined by the first forward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the forward transfer-transistor, the forward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second forward delay time determined by the second forward delay element to the forward capacitor, and, after the signal charge stored in the backward capacitor has been discharged, the backward transfer-transistor becomes active delayed by a first backward delay time determined by the first backward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the backward transfer-transistor, the backward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second backward delay time determined by the second backward delay element to the backward capacitor.
The bidirectional-marching memory of claim 12, wherein the first forward delay time and the first backward delay time are a quarter of clock period of the first clock signal, respectively, and the second forward delay time and the second backward delay time are a half of the clock period, respectively.
The bidirectional-marching memory of claim 10, wherein in the forward transfer-transistor and the backward transfer-transistor, the control-electrode controls a current flowing between the first main-electrode and the second main-electrode electro-statically.
The bidirectional-marching memory of claim 10, wherein in the forward reset-transistor and backward reset-transistor, the control-electrode controls a current flowing between the first main-electrode and the second main-electrode electro-statically.
The bidirectional-marching memory of claim 10, wherein the forward transfer-transistor, the forward reset-transistor, the backward transfer-transistor and the backward reset-transistor are made of an insulated-gate transistor, respectively, including a MOS transistor, a MIS transistor and a high electron mobility transistor.
The bidirectional-marching memory of claim 16, wherein the forward transfer-transistor, the forward reset-transistor, the backward transfer-transistor and the backward reset-transistor are made of an nMOS transistor, respectively, and the first clock signal of positively high-level is applied to the control electrode of the nMOS transistor so as to achieve a conductive state.
The bidirectional-marching memory of claim 16, wherein the forward transfer-transistor, the forward reset-transistor, the backward transfer-transistor and the backward reset-transistor are made of a pMOS transistor, respectively, and the first clock signal of negatively high-level is applied to the control electrode of the nMOS transistor so as to achieve a conductive state.
A bidirectional-marching memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential; and
a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.
The bidirectional-marching memory of claim 19, wherein in each of the bit-level cells, when a first clock signal from a first clock signal supply line is applied to the control-electrode of the forward reset-transistor, the forward reset-transistor discharges the signal charge already stored in the common capacitor, and a second clock signal from a second clock signal supply line is applied to the control-electrode of the backward reset-transistor, the backward reset-transistor discharges the signal charge already stored in the common capacitor.
The bidirectional-marching memory of claim 19, wherein in each of the bit-level cells, after the signal charge stored in the common capacitor has been discharged, the forward transfer-transistor becomes active delayed by a first forward delay time determined by the first forward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the forward transfer-transistor, the forward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second forward delay time determined by the second forward delay element to the common capacitor, and, after the signal charge stored in the common capacitor has been discharged, the backward transfer-transistor becomes active delayed by a first backward delay time determined by the first backward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the backward transfer-transistor, the backward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second backward delay time determined by the second backward delay element to the common capacitor.
A complex marching memory comprising a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size,
wherein each of the memory units transfers synchronously with a clock signal, step by step, toward an output side of corresponding marching memory block from an input side of the corresponding marching memory block, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.
The complex marching memory of claim 22, wherein each of the bit-level cells comprises:
a transfer-transistor having a first main-electrode connected to a clock signal supply line, configured to supply the clock signal, through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of memory units, through a second delay element;
a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and
a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor,
wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of memory units.
The complex marching memory of claim 23, wherein in each of the bit-level cells, when the clock signal is applied to the control-electrode of the reset-transistor, the reset-transistor discharges the signal charge already stored in the capacitor.
The complex marching memory of claim 23, wherein in each of the bit-level cells, after the signal charge stored in the capacitor has been discharged, the transfer-transistor becomes active delayed by a first delay time determined by the first delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the transfer-transistor, the transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second delay time determined by the second delay element to the capacitor.
A complex marching memory comprising a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size,
wherein each of the memory units transfers synchronously with a first clock signal, step by step, toward a first edge side of corresponding marching memory block from a second edge side of the corresponding marching memory block opposing to the first edge side, and further, each of the memory units transfers synchronously with a second clock signal, step by step, toward the second edge side from the first edge side, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.
The complex marching memory of claim 26, wherein each of the bit-level cells comprises:
a forward transfer-transistor having a first main-electrode connected to a first clock signal supply line, configured to supply the first clock signal, through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the first clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a second clock signal supply line, configured to supply the second clock signal, through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the second clock signal supply line, and a second main-electrode connected to the ground potential; and
a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.
The complex marching memory of claim 27, wherein in each of the bit-level cells, when the first clock signal is applied to the control-electrode of the forward reset-transistor, the forward reset-transistor discharges the signal charge already stored in the common capacitor, and the second clock signal is applied to the control-electrode of the backward reset-transistor, the backward reset-transistor discharges the signal charge already stored in the common capacitor.
The complex marching memory of claim 27, wherein in each of the bit-level cells, after the signal charge stored in the common capacitor has been discharged, the forward transfer-transistor becomes active delayed by a first forward delay time determined by the first forward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the forward transfer-transistor, the forward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second forward delay time determined by the second forward delay element to the common capacitor, and, after the signal charge stored in the common capacitor has been discharged, the backward transfer-transistor becomes active delayed by a first backward delay time determined by the first backward delay element, and when the signal stored in the first neighboring bit-level cell is fed to the control-electrode of the backward transfer-transistor, the backward transfer-transistor transfers the signal stored in the first neighboring bit-level cell, further delayed by a second backward delay time determined by the second backward delay element to the common capacitor.
A computer system comprising a processor and a marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the marching main memory to the processor, the marching main memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a transfer-transistor having a first main-electrode connected to a clock signal supply line through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of the memory units, through a second delay element;
a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and
a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor,
wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of the memory units.
A computer system comprising a processor and a bidirectional marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the bidirectional marching main memory to the processor, the bidirectional marching main memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a forward transfer-transistor having a first main-electrode connected to a forward clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the forward clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a backward clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the backward clock signal supply line, and a second main-electrode connected to the ground potential; and
a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.
A computer system comprising a processor and a bidirectional marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the bidirectional marching main memory to the processor, the bidirectional marching main memory including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size, each of the bit-level cells comprising:
a forward transfer-transistor having a first main-electrode connected to a forward clock signal supply line through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of the memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the forward clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a backward clock signal supply line through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the backward clock signal supply line, and a second main-electrode connected to the ground potential; and
a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of the memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.
A computer system comprising a processor and a marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the marching main memory to the processor, the marching main memory comprising a plurality of marching memory blocks being deployed spatially, each of the marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells configured to store information of byte size or word size,
wherein each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.
The computer system of claim 33, wherein each of the bit-level cells comprises:
a transfer-transistor having a first main-electrode connected to a clock signal supply line through a first delay element and a control-electrode connected to an output terminal of a first neighboring bit-level cell disposed at input side of the array of memory units, through a second delay element;
a reset-transistor having a first main-electrode connected to a second main-electrode of the transfer-transistor, a control-electrode connected to the clock signal supply line, and a second main-electrode connected to the ground potential; and
a capacitorconfigured to store the information of the bit-level cell, connected in parallel with the reset-transistor,
wherein an output node connecting the second main-electrode of the transfer-transistor and the first main-electrode of the reset-transistor serves as an output terminal of the bit-level cell, and the output terminal of the bit-level cell delivers the signal stored in the capacitor to a second neighboring bit-level cell disposed at output side of the array of memory units.
A computer system comprising a processor and a bidirectional marching main memory configured to provide the processor with stored information actively and sequentially so that the processor can execute arithmetic and logic operations with the stored information, in addition results of processing in the processor are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is only one way of instructions flow from the bidirectional marching main memory to the processor, the bidirectional marching main memory comprising a plurality of bidirectional marching memory blocks being deployed spatially, each of the bidirectional marching memory blocks including an array of memory units, each of the memory units having a sequence of bit-level cells so as to store information of byte size or word size,
wherein each of the memory units transfers synchronously with a first clock signal, step by step, toward a first edge side of corresponding marching memory block from a second edge side of the corresponding marching memory block opposing to the first edge side, and further, each of the memory units transfers synchronously with a second clock signal, step by step, toward the second edge side from the first edge side, and each of the marching memory blocks is randomly accessed so that each of the memory units in a subject marching memory block can be randomly accessed.
The computer system of claim 35, wherein each of the bit-level cells comprises:
a forward transfer-transistor having a first main-electrode connected to a forward clock signal supply line, configured to supply the first clock signal, through a first forward delay element and a control-electrode connected to an forward output terminal of a first neighboring bit-level cell disposed at a one side of the array of memory units, through a second forward delay element;
a forward reset-transistor having a first main-electrode connected to a second main-electrode of the forward transfer-transistor, a control-electrode connected to the forward clock signal supply line, and a second main-electrode connected to the ground potential;
a backward transfer-transistor having a first main-electrode connected to a backward clock signal supply line, configured to supply the second clock signal, through a first backward delay element and a control-electrode connected to an backward output terminal of the second neighboring bit-level cell through a second backward delay element;
a backward reset-transistor having a first main-electrode connected to a second main-electrode of the backward transfer-transistor, a control-electrode connected to the backward clock signal supply line, and a second main-electrode connected to the ground potential; and
a common capacitorconfigured to store the information of the bit-level cell, connected in parallel with the forward reset-transistor, and the backward reset-transistor,
wherein an output node connecting the second main-electrode of the forward transfer-transistor and the first main-electrode of the forward reset-transistor serves as an forward output terminal of the bit-level cell, the forward output terminal of the bit-level cell delivers the signal stored in the common capacitor to a second neighboring bit-level cell disposed at another side of the array of memory units, an output node connecting the second main-electrode of the backward transfer-transistor and the first main-electrode of the backward reset-transistor serves as a backward output terminal of the bit-level cell, and the backward output terminal of the bit-level cell delivers the signal stored in the common capacitor to the first neighboring bit-level cell.
The computer system of claim 36, wherein in each of the bit-level cells, when the first clock signal is applied to the control-electrode of the forward reset-transistor, the forward reset-transistor discharges the signal charge already stored in the common capacitor, and the second clock signal is applied to the control-electrode of the backward reset-transistor, the backward reset-transistor discharges the signal charge already stored in the common capacitor.
The computer system as in any of claims 30-32, 34 or 36, wherein the processor further includes a plurality of arithmetic pipelines configured to receive the stored information from the bidirectional marching main memory.
The computer system as in any of claims 30-32, 34 or 36, further comprising a marching-cache memory having an array of cache memory units, located at locations corresponding to each for a unit of information, cache input terminals of the array configured to receive the stored information from the bidirectional marching main memory, and cache output terminals of the array, configured to store information in each of cache memory units and to transfer, synchronously with the clock signal, step by step, the information each to an adjacent cache memory unit, so as to provide actively and sequentially the stored information to the processor so that the arithmetic logic unit can execute the arithmetic and logic operations with the stored information, the results of the processing in the arithmetic logic unit are sent out to the bidirectional marching main memory, except that in case of instructions movement, there is not the opposite direction to the information flow to be processed..
The computer system of claim 39, wherein the plurality of arithmetic pipelines includes either a plurality of vector processing units or a plurality of scalar function units.
The computer system as in any of claims 30-32, 34 or 36, further comprising a marching-cache memory having an array of cache memory units, cache input terminals of the second array configured to receive the stored information from the bidirectional marching main memory, and cache output terminals of the second array, configured to store information in each of cache memory units and to transfer successively, step by step, the stored information in each of cache memory units to an adjacent cache memory unit, synchronized with the clock signal from the cache memory units adjacent to the cache input terminals toward the cache memory units adjacent to the cache output terminals, so as to provide actively and sequentially the stored information to the processor through the cache output terminals so that the processor cores can operate with the stored information.