WO2001098880A1 - Systeme de memoire rapide - Google Patents
Systeme de memoire rapide Download PDFInfo
- Publication number
- WO2001098880A1 WO2001098880A1 PCT/JP2001/005270 JP0105270W WO0198880A1 WO 2001098880 A1 WO2001098880 A1 WO 2001098880A1 JP 0105270 W JP0105270 W JP 0105270W WO 0198880 A1 WO0198880 A1 WO 0198880A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- memory
- data
- circuit
- switch
- memory system
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1051—Data output circuits, e.g. read-out amplifiers, data output buffers, data output registers, data output level conversion circuits
- G11C7/1057—Data output buffers, e.g. comprising level conversion circuits, circuits for adapting load
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1668—Details of memory controller
- G06F13/1684—Details of memory controller using multiple buses
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1006—Data managing, e.g. manipulating data before writing or reading out, data bus switches or control circuits therefor
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1051—Data output circuits, e.g. read-out amplifiers, data output buffers, data output registers, data output level conversion circuits
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1051—Data output circuits, e.g. read-out amplifiers, data output buffers, data output registers, data output level conversion circuits
- G11C7/106—Data output latches
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1051—Data output circuits, e.g. read-out amplifiers, data output buffers, data output registers, data output level conversion circuits
- G11C7/1066—Output synchronization
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1078—Data input circuits, e.g. write amplifiers, data input buffers, data input registers, data input level conversion circuits
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1078—Data input circuits, e.g. write amplifiers, data input buffers, data input registers, data input level conversion circuits
- G11C7/1084—Data input buffers, e.g. comprising level conversion circuits, circuits for adapting load
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
- G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
- G11C7/1078—Data input circuits, e.g. write amplifiers, data input buffers, data input registers, data input level conversion circuits
- G11C7/1087—Data input latches
Definitions
- the present invention relates to a high-speed memory system, and more particularly, to a memory system having a high transmission speed and a high-speed operation, and a memory interface and a memory chip used in the memory system.
- Figure 18 shows a conventional memory system.
- a bus 1802 connected to a memory controller 1801 extends in a direction, and a plurality of (two in the figure) memory modules 1804 mounted with a plurality of (two in the figure) memory chips They are connected in parallel.
- a connection structure to a bus that is, a memory interface is referred to as a bus connection type interface represented by a synchronous DRAM (SDRAM) or a rambus DRAM (Rambus DRAM, RDRAM). Called.
- SDRAM synchronous DRAM
- Rambus DRAM Rasterbus DRAM
- the bus-connected interface has a great advantage in terms of memory expandability because multiple memory chips or memory modules are connected in parallel to an extended bus.
- the transmission speed seems to be limited to about 1-2 Gb / s.
- the data queue caused by the difference in transmission line length can be ignored. It may disappear. That is, the phase of the data input to the memory differs for each data, and it is impossible to simultaneously capture data that should be input at the same time. As a result, a malfunction may occur. This becomes remarkable as the data transmission speed increases.
- FIG. 19 is a block diagram showing a conventional memory chip.
- an 8-bit command / address packet input serially from the bus is converted into a parallel signal by 1: 8 DEMUX (serial / parallel conversion circuit) 1901, and then decoded by a packet decoding circuit (decoder) 1902.
- the memory cells are input to the memory core 1904 in which the memory cells are arranged in a matrix.
- serial 8-bit input data is converted into parallel data by the 1: 8 DEMUX 1905 to become 64-bit data, input to the memory core 1904, and output from the memory core 1904 as parallel 64-bit output data.
- serial conversion is performed by 8: 1 MUX 1906 and output as 8-bit serial data.
- FIG. 20 is a block diagram showing the input data fetching in the conventional memory chip.
- the input data and the 1: 8 DEM UX of the conventional input / output data shown in FIG. 19 are described in detail.
- the input data width is 8 bits
- the input data 2001 is input on an 8-bit bus
- each data 200 1 A to 2001H is 8 1: 8 D EMUX. Entered in 2002 A ⁇ 2002 H.
- These 8-bit input data 200 1 A to 200 1H are taken into 1: 8 DEMUX 2002A to 2002H by one clock input 20004.
- the clock 2004 captures the data 2001A to 2001H at the same timing at the transition point of the data 2001A to 2001H and almost at the center of the transition point (indicated by the dotted line).
- the 8-bit input data 2001A to 2001H has a slight difference in the data phase due to the difference in the length of the transmission path through which the input data is transmitted. This causes a gap. This is called data skew 2 0 10. If the data skew 21010 is negligibly small compared to the data rate, 8-bit data capture by one clock 204 is performed normally.
- the data skew cannot be ignored, that is, if the data transmission rate becomes faster and the data skew cannot be ignored compared to the data rate, all the 8-bit data is captured at the same timing with one clock. You will not be able to do it. That is, one of the limiting factors of the transmission speed of transmission data in the conventional memory chip is, as described above, a plurality of data including data skew caused by a difference in transmission path length or the like. Cannot be captured at the same evening.
- the first problem of the prior art is that each memory chip or memory module is connected to each part of the extended bus, so that the reflection on the transmission line is large and the load (fan-out) on the transmission line is large. Therefore, it is difficult to increase the transmission speed on the bus.
- a second problem with the prior art is that, because a memory chip or a memory module is connected to each part of the extended bus, data queues caused by differences in transmission line length cannot be ignored. .
- the third problem of the prior art is that since no effective countermeasure against the data skew is taken in the memory chip, there is a concern that a malfunction may occur. Disclosure of the invention
- an object of the present invention is to provide a memory system and a memory interface that realize a higher transmission speed by suppressing reflection and load on a transmission line.
- Another object of the present invention is to provide a memory system and a memory interface in which data skew caused by a difference in transmission line length is suppressed.
- Another object of the present invention is to provide a memory chip having an effective measure against data skew.
- Still another object of the present invention is to provide a memory system using a memory chip having an effective countermeasure against data skew and having an increased transmission speed.
- the present invention provides a memory system having one memory controller and a plurality of memories, wherein the plurality of memories are connected to predetermined locations of a bus connected to the memory controller via a switch, There is provided a memory system in which the operation of the plurality of memories is controlled by the switch.
- the switches are arranged hierarchically. Further, it is preferable that the switch is a switch of a single type or a switch of MUX (parallel / serial conversion) ZDEMUX (serial / parallel conversion).
- the switch is provided with a data recovery circuit. It is also preferable that a data recovery circuit is provided in either the memory controller or the memory.
- a plurality of memories may be formed on individual memory chips, respectively, and the switches may be formed on individual switch chips.
- the memory chips and the switch chips are mounted on the same memory module.
- a plurality of the memory modules may be connected to the memory controller through individual paths.
- a memory system having one memory controller and a plurality of memories, wherein the plurality of memories are respectively connected to the memory controller by individual buses.
- a recovery circuit may be provided in either the memory controller or the memory.
- a memory interface provided in a memory controller, the data interface including a data recovery circuit connected to a bus, and a switch connecting a plurality of memories to a predetermined portion of the path. .
- the switch is provided with a data recovery circuit.
- a memory chip in which a command / address signal and a data signal are input to a memory core in which a plurality of memory cells are arranged in a matrix, wherein the command / address signal is provided. And each of the data signals provides a memory chip input through a recovery circuit.
- Each of the command / address signal and the data signal can be input to the memory core through a data synchronization circuit and a data synchronization circuit and a bucket combining circuit.
- a DEMUX (serial-parallel conversion) circuit may be provided between the data recovery circuit and the data synchronization circuit.
- the individual memories are connected to the memory controller in a one-to-one relationship, so that the reflection on the transmission line increases and the load (fan-out) increases. Therefore, it is possible to increase the transmission speed on the bus.
- FIG. 1 is a schematic arrow view of a memory system according to an embodiment of the present invention.
- FIG. 2 is a schematic arrow view showing a modification of the memory system of FIG.
- FIG. 3 is a block diagram showing a switch of the memory system of FIG.
- FIG. 4 is a block diagram showing another example of the switch of the memory system of FIG.
- FIG. 5 is a block diagram showing a configuration of a memory chip of the memory system of FIG. 1.
- FIG. 6 is a timing chart of input data in the memory chip of FIG. 5.
- FIG. 7 is a timing chart of output data of the memory chip of FIG. FIG. 8, which is a timing chart, is an evening timing chart of the command / address bucket at each position in FIGS. 3 and 5.
- FIG. 9 is a block diagram showing the capture of input data of the memory system of FIG.
- FIG. 10 is an evening timing chart showing signals at respective positions in FIG.
- FIG. 11 is a block diagram of a memory controller in the memory system of FIG.
- FIG. 12 is an evening chart of the memory controller of FIG.
- FIG. 13 is a block diagram of a 1: 2 serial / parallel conversion circuit.
- Figure 14 is a block diagram of a 1: 8 serial to parallel conversion circuit.
- FIG. 15 is a block diagram of the data synchronization circuit.
- FIG. 16 is a block diagram of the data recovery circuit.
- FIG. 17 is a block diagram of a memory system according to another embodiment of the present invention.
- FIG. 18 is a schematic arrow view of a conventional memory system.
- FIG. 19 is a block diagram showing a memory chip of the memory system of FIG.
- FIG. 20 is a block diagram showing the capture of input data of the memory chip of FIG.
- FIG. 21 is a timing chart of each position in FIG. Preferred embodiments of the invention The present invention will be described below with reference to the drawings.
- a memory system according to an embodiment of the present invention includes a memory controller 101 and a plurality of memories 104, and a predetermined bus 105 connected to the memory controller 101.
- a plurality of memories 104 are connected to the end portion, which is the portion via the switch 103 and the wiring means 105.
- the operation of the plurality of memories 104 is controlled by the switch 103.
- the plurality of memories 104 are formed on individual memory chips 104, respectively, and the switches 103 are formed on individual switch chips 103. These chips are mounted on the same memory module 102.
- a large number of memories are connected by connecting the plurality of memory modules 102 to the memory controller 101 via individual buses 106, respectively.
- Each bus 106 includes a wiring group such as a data signal line and a control signal line.
- the number of signal lines for transmitting n-bit data in parallel is n
- the number of signal lines for transmitting n-bit data in serial is one.
- the case of serial transmission will be described as an example. However, the configuration and effect of the present invention are the same in the case of parallel transmission.
- the operation of a plurality of memories 104 that is, memories A and B is controlled by switches, so that even if memory B is added to memory A,
- This memory has a connection relationship between the memory controller 101 and 1 enclosure.Therefore, even if the number of memories is increased, the reflection on the transmission line does not increase or the number of loads (fans) does not increase. Therefore, the transmission speed on the bus can be increased.
- each memory module 1 0 2 is connected to the memory controller 101 by a separate bus 106, so that there is no problem in increasing the transmission speed.
- FIG. 2 shows a modified example of the memory system of FIG. 1, and shows an example in which the inside of a memory module is changed.
- FIG. 2 the same or similar elements as those in FIG. 1 are denoted by the same reference numerals, and overlapping description will be omitted.
- the memory module 102 in FIG. 2 shows that the number of memories is increased by hierarchically connecting the switches 103.
- switch 103 is connected to two layers. Also in this case, since each memory 104 has a one-to-one connection with the memory controller 101 via two switches 103, the reflection on the transmission line is large even if the memory is increased. The transmission load on the bus does not increase and the transmission speed on the bus can be increased.
- FIG. 3 is a block diagram showing a star-type switch as an example of the switch 103 of FIGS.
- the signal flow in the direction from the memory controller to the memory via the bus is constituted by 301 to 313 in FIG.
- the signal flow in the direction from the memory to the memory controller via the bus consists of 314 to 319 in FIG.
- the command and address packets are taken into the chip by the data recovery circuit 301.
- the recovered packet is parallel-converted using a serial-to-parallel conversion circuit (for example, in the case of 8-bit, 1: 8 DEMUX) 302.
- the number of bits to be parallel-converted differs depending on the specification, and may not be parallel-converted.
- the phase of the parallel signal may be different due to the difference in the length of the signal medium outside the chip and the difference in the internal clock state of the serial / parallel conversion circuit 302.
- the data synchronization circuit 303 aligns the phases of the parallel signals. Commands and addresses with the same phase
- the packet is decrypted by the bucket decryption circuit 304, and the bucket decryption circuit 304 outputs the command and address bucket converted in parallel and the decryption result 313.
- the command and address bucket subjected to parallel conversion are converted into serial signals by a parallel / serial conversion circuit (for example, 8: 1 MUX in the case of 8 bits) 305.
- a parallel / serial conversion circuit for example, 8: 1 MUX in the case of 8 bits
- the number of parallel bits subjected to serial conversion is the same as the number of bits parallelized by the serial / parallel conversion circuit 302.
- the number of parallel bits is not limited to 8, but may be 10 bits or 16 bits, and the parallel-serial conversion circuit 305 may not be used.
- the serial-converted bucket outputs a command Z address signal to any memory or the next-stage switch by, for example, a CMOS configuration switch element 360. Is determined. For example, it is determined whether signal A is transmitted to memory A and signal B is not transmitted to memory B, or conversely, signal A is not transmitted to memory A and signal B is transmitted to memory B.
- Data from the memory controller is more and more taken into the chip by the delay recovery circuit 307.
- the recovered data is parallel-converted by the serial / parallel conversion circuit 308.
- the number of bits to be parallel-converted is not limited to eight pits, and the serial-parallel conversion circuit 308 may not be used.
- the data synchronization circuit 310 uses the data synchronization signal 310 to equalize the phase difference.
- the parallel-to-parallel converted data having the same phase is subjected to serial conversion by the parallel / serial conversion circuit 311 by the number of parallel bits converted by the serial / parallel conversion circuit 3108.
- the serial-converted data is transferred to any memory or the next-stage switch according to the packet decoding result 3 13, for example, by the switch element 3 12 of the CMOS configuration. It is decided whether to convey. example For example, whether to transmit data A to memory A and not transmit data B to memory B, or vice versa, do not transmit data A to memory A and transmit data B to memory B Is determined.
- the data flow in the direction from the memory to the memory controller is as follows.
- the data recovery circuit 319 fetches data from the memory chips A and B.
- the recovered data is parallel-converted by the serial / parallel conversion circuit 318.
- the number of bits to be parallel-converted is not limited to 8 bits, and there are some which are not used by the serial / parallel conversion circuit 318.
- the parallel-converted data is converted by the data synchronization circuit 3 17 using the data synchronization signal 3 20.
- the phases are aligned.
- the data having the same phase is determined by the control signal 3 21 from the bucket decoding circuit 3 16 from the memory 3 or the switch 3 15 in the CMOS configuration, which memory or switch should be passed to the memory controller. Selected. For example, select whether to pass data A from memory A and not pass data B from memory B, or vice versa, to pass data B from memory B without passing data A from memory A. I do.
- the selected data is serial-converted by the parallel / serial conversion circuit 314 and output.
- the number of parallel data bits subjected to serial conversion is the same as the number of bits subjected to parallel conversion by the serial-parallel conversion circuit 318.
- FIG. 4 shows a MUX / DEMUX type switch which is another example of the switch 103 of FIGS.
- the flow in the direction from the memory controller to the memory via the bus is constituted by 401 to 413 in FIG.
- the signal flow in the direction from the memory to the memory controller via the bus is indicated by 414 to 418 in FIG.
- command The addressless bucket is taken into the chip by the data recovery circuit 401.
- the recovered packet is subjected to parallel conversion using a serial / parallel conversion circuit (for example, 1: n DEMUX) 402 in the case of n bits.
- the number of bits to be parallel-converted differs depending on the specification, and may not be parallel-converted.
- the phase of the parallel signal may be different due to the difference in the length of the signal medium outside the chip or the difference in the internal clock state of the serial / parallel conversion circuit 402.
- the data synchronization circuit 403 aligns the phases of the parallel signals using the data synchronization signal 408 input from the outside.
- the command and address packet having the same phase are decoded by the packet decoding circuit 404, and the packet decoding circuit 404 outputs the command and address packet converted in parallel and the decoding result 410.
- the signal of the decoding result 410 controls, for example, the switch element 405 having a CMOS configuration to determine which command or address signal is transmitted to which memory or the next-stage switch. For example, it is determined whether signal A is transmitted to memory A and signal B is transmitted to memory B, or conversely, signal A is transmitted to memory B and signal B is transmitted to memory A.
- Data from the memory controller is increasingly taken into the chip by the data recovery circuit 406.
- the recovered data is subjected to parallel conversion by a serial / parallel conversion circuit 407.
- the number of bits to be parallel-converted is not limited to 8 bits, and the serial-parallel conversion circuit 407 may not be used.
- the data synchronization circuit 409 uses the data synchronization signal 408 to make the phase difference uniform.
- the data that has been phase-aligned and converted into parallel data can be stored in any memory by, for example, a CMOS-structured switch element 413 according to the packet decryption result by the packet decryption circuit 412. Or, which data is transmitted to the next-stage switch is determined. For example, it is determined whether data A is transmitted to memory A and data B is transmitted to memory B, or conversely, data A is transmitted to memory B and data B is transmitted to memory A.
- the flow of data in the direction from the memory to the memory controller is as follows. First, the data is read from the memory chips A and B by the data recovery and retransmission circuit 416. Since the recovered data may have a phase difference similar to the data from the memory controller, the data is synchronized by the data synchronization circuit 415 using the data synchronization signal 418. Can be The data with the same phase is determined by the parallel / serial conversion circuit 414, which data from the memory or the switch should be passed to the memory controller and in which order.
- FIG. 5 shows the memory chip 104 of FIGS. 1 and 2—an example.
- a command and an address packet input from the memory controller or the switch are input to the inside of the chip by using the digital force validating circuit 501.
- the recovered command and address bucket output from the data recovery circuit 501 are converted into parallel data by the serial / parallel conversion circuit 502.
- a 1: 8 serial / parallel conversion circuit (1: 8 DEMUX) 502 for converting to 8-bit parallel data is used.
- the number of bits of parallel data may vary depending on the command and address bucket specifications, and may be 10 pits or 16 bits.
- the serial-to-parallel conversion circuit 502 may not be used depending on the packet specifications.
- the phase of the command and address packet after the parallel conversion depends on the difference in the length of the transmission medium between the switch and the memory and the clock state of the serial / parallel conversion circuit 502. Dependent and may be different. Therefore, the phases of the command and the address bucket subjected to the parallel conversion are aligned by the data synchronization circuit 503 using the data synchronization signal 500 input from the outside.
- the command and the address packet whose phases have been aligned are decoded by the packet decoding circuit (decoder) 504 and transmitted to the memory core 505 in which memory cells are arranged in a matrix. . '
- the input data 513 is taken into the chip by the data recovery circuit 506, and then converted into parallel data by the serial / parallel conversion circuit 507.
- the number of bits of the parallel data depends on the specification of the memory data.
- an example of 8 bits is shown, but there are also 10 bits and 16 bits.
- the serial / parallel conversion circuit 507 may not be used depending on data specifications.
- the phase of the parallel-converted data is aligned by the data synchronization circuit 509.
- the data synchronization circuit 509 aligns the phases of the data using an external data synchronization signal 500 as in the data synchronization circuit 503 for command and address buckets.
- the data whose phases have been aligned are decoded by the packet decoding circuit 510 and input to the memory core 505.
- the bucket decoding circuit 510 is not used depending on the specifications of the database.
- the output data is converted from, for example, 8-bit parallel data from the memory core 505 into a serial data by a parallel / serial conversion circuit 508 and output.
- the number of bits of the parallel data to be serially converted is determined by the data specification, and may be output in 10 bits, 16 bits, or without serial conversion.
- a packet encoding circuit 5111 may be inserted between the memory core 505 and the parallel-to-serial conversion circuit 508.
- FIG. 6 is a timing chart of input data at each location of the memory chip of FIG.
- FIG. 7 is a timing chart of output data at each point of the memory chip of FIG. Fig. 8 shows the command nodes at each point in Figs. 3 and 5. Timing chart.
- Input data (Fig. 6): The input data from 5 13 [0] to 5 13 [7] has a low data queue due to the difference in the length of the data transmission path between the memory controller and the memory chip. May have.
- a plurality of input data having a data queue by the data recovery circuit 506 are fetched into a chip at a clock 515 at an optimal timing for each data, and passed to the DE MUX circuit 507. 5 14.
- the data skew is also transferred to the output 516 and the clock 517 of the 1: 8 DEMUX 507.
- the data queue synchronizes the output 516 of the DEMUX 507 with the internal clock 520 of the memory chip by the data synchronization circuit 509 using the data synchronization signal 500.
- the input data is decoded by the packet decoding circuit 510 and input to the memory core 505 519.
- Data 521 from data core 505 is synchronized with internal clock 520.
- the data is encoded 522 by a bucket encoding circuit 511, and a data synchronization signal 512 is generated in synchronization with the header at the head of the bucket.
- the encoded data 522 is input to the 8: 1 MUX 508, converted into a serial signal using the high-speed clock in the MUX circuit, and output 523.
- FIG. 8 showing a timing chart of a command address packet in the memory chip or the switch chip of the present invention will be described.
- the bit width of the command Z address bucket is 8 bits as an example.
- the input command / address packet may have a data queue due to a difference in the length of the data transmission path between the memory controller and the switch chip or between the switch chip and the memory chip.
- Multiple input commands with data queue Address packets are processed by data recovery circuits 501 and 301, respectively. 5 24, 3 24, which are captured on the chip at the clock with the best timing for these command / address packets and passed to the DEMUX circuits 502, 302.
- this data skew is also transferred to the outputs 525, 325 of the 1: 8 DEMUX 502, 302.
- this data skew is output by the data synchronization circuits 503, 303 using the data synchronization signals 50,000, 310, to output the DEMUXs 502, 302, 525, 322. Synchronize 5 with the clock 5 2 0 inside each chip 5 2 6, 3 2 6 contend
- FIG. 9 is a block diagram relating to the capture of input data in the memory chip of the present invention
- FIG. 10 is a timing chart thereof.
- the memory chip of the present invention solves the data capture error caused by the conventional data skew described with reference to FIGS. 20 and 21 and increases the data rate, that is, the data transmission speed. can do.
- the data recovery circuit 902 is a circuit which takes in data input to the data recovery circuit at the optimum timing for the data. Therefore, even if there is a data skew, a plurality of data input to the memory chip can always be taken into the chip in 1-bit data units.
- the data captured in the chip by the data recovery circuit has a different phase for each data. That is, 903A to 903H are output using clocks 904A to 904H at the optimum timing for the input data.
- These data 903 and clock 904 are input to eight 1: 8 DEMUX 905 A to 905 H, respectively. Since the 1: 8 DEMUX operates according to the input clock, the eight 1: 8 DEMUXs 905A to 905H operate in synchronization with clocks having independent phases.
- the phase of the 906 H is also independent between the eight .D EMUX. Therefore, in order to align the data having these disparate phases to one phase, the output 906 of the DEMUX is input to the data synchronization circuit 908, and the synchronization between the outputs 906A to 906H of the DEMUX is performed. Outputs data 909.
- a data recovery circuit is provided for each bit of input data, and each data is taken into the chip at an optimum timing. Therefore, the problem of data skew, which has been a limiting factor of the data transmission speed in the conventional memory chip, can be solved, and the memory chip of the present invention enables data transmission at a higher speed than the conventional memory chip.
- a data recovery circuit 902 connected to a path to suppress adverse effects of data skew due to a difference in transmission path length, and a 1: 8 DEMUX 905 and a data synchronization circuit connected thereto are connected.
- the circuit 908 can be considered to form part of the memory interface.
- FIG. 11 is a block diagram illustrating the memory controller of the embodiment, and FIG. 12 is a timing chart thereof.
- Address input is through address buffer 1101, address decoding circuit 1102, timing control and packet encoding and decoding circuit 1103, multiplexer 1104, buffer 1105, 8: 1 MUX 1106
- a command signal input to a switch element 1107 having a CMOS configuration, the timing control and packet coding, and a control signal 1108 from a decoding circuit 1103 activates the switch element 1107 to select a command node.
- the address bucket is output on the path.
- the data input is through the data buffer 111, timing control and packet coding, decoding circuit 1103, multiplexer 1111, buffer 1 112, 8: 1 MUX 111 13
- the data is input to the switch element 1 114 having the S configuration, and the data selected by the switch element 1 114 being operated by the control signal 1 108 from the timing control and packet coding / decoding circuit 1103 is passed to the path.
- the data output from the memory is taken into the data recovery circuit 1118 via the bus, and 1: 8 DEMUX 1117, data synchronization circuit 1116, evening imaging control and packet coding and decoding circuit 1103, Data is output to the CPU through the buffer 1115.
- the data recovery circuit, DEMUX, and data synchronization circuit on the input side (the side receiving the output data from the memory) of the memory controller shown in Fig. 11, input (output from the memory) Even if there is a phase difference, the phases are aligned, thereby suppressing malfunction due to data skew and enabling high-speed operation.
- the data recovery circuit 1 118 and its connection 1: 8 DEMUX 1 117 and data synchronization circuit 1 1 16 Constitutes a part of the memory interface.
- Figure 13 shows a block diagram of a 1: 2 serial / parallel conversion circuit (1: 2 D EMUX). It comprises a master 'slave-type flip-flop 1302 comprising latches 1306 and 1307, and a master' slave-master-type flip-flop 1301 comprising latches 1303-1305. It has a function to convert data into two at the rising edge and falling edge of the clock at half the speed of the input data.
- Figure 14 shows the block diagram of the 1: 8 serial / parallel converter.
- a 1: 8 serial-to-parallel conversion circuit can be obtained by connecting a 1: 2 serial-to-parallel conversion circuit 1400 to 1406 in a tree structure.
- the clocks distributed to each stage are produced by using a divide-by-2 divider 1407-1409.
- FIG. 15 is a circuit diagram showing a data synchronization circuit.
- the data synchronization circuit is composed of a header detection circuit 1501 and a data extraction circuit 1502. By adjusting the timing of the data synchronization part and the timing of the data synchronization signal input from the outside, the data synchronization is performed in a short time. It is a configuration that takes synchronization.
- a data synchronization circuit that requires an external data synchronization signal has been described, but it is also possible to encode data and include the data synchronization signal in input data. When such encoded data is used, data synchronization can be performed without the need for an external data synchronization signal.
- FIG 16 shows the configuration of the data recovery circuit.
- the phase difference circuit detects the phase difference between the input clock and the internal clock using the phase comparison circuit 1601, and determines whether the data is advanced or delayed with respect to the internal clock. Outputs phase signal and late signal.
- the delay control circuit 1602 changes the phase of the clock according to the comparison result of the phase comparison circuit 1601, and adjusts the phase of the internal clock to the input data.
- the input data is waveform-shaped by flip-flop 163 or the like, and data 164 is output. It also outputs the phase-matched internal clock 1605.
- FIG. 17 is a diagram showing a memory system according to another embodiment of the present invention.
- a plurality of memory chips 104 are connected to the memory controller 101 by individual buses 106, respectively.
- each memory chip 104 is connected to the memory controller 101 on a one-to-one basis, even if the memory is increased, the reflection on the transmission line increases or the load (fan-out) increases. Since the transmission speed does not increase, it is possible to increase the transmission speed on the path.
- a memory system in which each memory chip is connected to a memory of a memory controller through an individual bus as shown in FIG. 17 ' is suitable.
- a memory system that connects the memory of a memory controller via a bus via a switch as shown in Fig. 1 is suitable.
- a memory system in which the switch in Fig. 1 is used as the MUX / DEMUX in Fig. 4 and the memory of the memory controller is connected via a bus via this switch is suitable.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Dram (AREA)
- Time-Division Multiplex Systems (AREA)
- Memory System (AREA)
- Logic Circuits (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/311,687 US7366821B2 (en) | 2000-06-21 | 2001-06-20 | High-speed memory system |
KR10-2002-7017505A KR100512895B1 (ko) | 2000-06-21 | 2001-06-20 | 고속메모리시스템 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2000186234A JP2002007201A (ja) | 2000-06-21 | 2000-06-21 | メモリシステム、メモリインターフェース及びメモリチップ |
JP2000-186234 | 2000-06-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2001098880A1 true WO2001098880A1 (fr) | 2001-12-27 |
Family
ID=18686430
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2001/005270 WO2001098880A1 (fr) | 2000-06-21 | 2001-06-20 | Systeme de memoire rapide |
Country Status (6)
Country | Link |
---|---|
US (1) | US7366821B2 (ja) |
JP (1) | JP2002007201A (ja) |
KR (1) | KR100512895B1 (ja) |
CN (2) | CN101561707B (ja) |
TW (1) | TW511085B (ja) |
WO (1) | WO2001098880A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100461085C (zh) * | 2002-09-18 | 2009-02-11 | 株式会社日立制作所 | 存储系统及其控制方法 |
Families Citing this family (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100518532B1 (ko) * | 2002-04-27 | 2005-10-04 | 삼성전자주식회사 | 선택적으로 명령신호 및 어드레스를 전송하는 방법 및 장치 |
JP4094370B2 (ja) | 2002-07-31 | 2008-06-04 | エルピーダメモリ株式会社 | メモリモジュール及びメモリシステム |
US7382803B1 (en) * | 2002-08-12 | 2008-06-03 | Broadcom Corporation | Hybrid high-speed/low-speed output latch in 10 GBPS interface with half rate clock |
GB2424105B (en) * | 2003-01-13 | 2007-03-07 | Rambus Inc | Coded write masking |
US8458201B2 (en) * | 2005-04-08 | 2013-06-04 | International Business Machines Corporation | Method and apparatus for mapping structured query language schema to application specific business objects in an integrated application environment |
US20060230048A1 (en) * | 2005-04-08 | 2006-10-12 | International Business Machines Corporation | Method and apparatus for object discovery agent based mapping of application specific markup language schemas to application specific business objects in an integrated application environment |
DE102005016684A1 (de) * | 2005-04-11 | 2006-10-12 | Deutsche Thomson-Brandt Gmbh | Speicheranordnung, insbesondere zur nichtflüchtigen Speicherung von unkomprmierten Video-und/oder Audiodaten |
US9582449B2 (en) | 2005-04-21 | 2017-02-28 | Violin Memory, Inc. | Interconnection system |
US9286198B2 (en) | 2005-04-21 | 2016-03-15 | Violin Memory | Method and system for storage of data in non-volatile media |
US8112655B2 (en) | 2005-04-21 | 2012-02-07 | Violin Memory, Inc. | Mesosynchronous data bus apparatus and method of data transmission |
CN101727429B (zh) * | 2005-04-21 | 2012-11-14 | 提琴存储器公司 | 一种互连系统 |
US8452929B2 (en) | 2005-04-21 | 2013-05-28 | Violin Memory Inc. | Method and system for storage of data in non-volatile media |
US9384818B2 (en) | 2005-04-21 | 2016-07-05 | Violin Memory | Memory power management |
US20070061494A1 (en) | 2005-08-30 | 2007-03-15 | Paul Wallner | Semiconductor memory system, semiconductor memory chip, and method of masking write data in a semiconductor memory chip |
TWI293733B (en) * | 2005-11-22 | 2008-02-21 | Novatek Microelectronics Corp | Asynchronous bus processing apparatus |
US20070239906A1 (en) * | 2006-03-13 | 2007-10-11 | Vakil Kersi H | Input/output agent having multiple secondary ports |
KR100801709B1 (ko) * | 2006-05-16 | 2008-02-11 | 삼성전자주식회사 | 메모리 모듈 및 이를 구비한 메모리 시스템 |
US8185711B2 (en) | 2006-05-16 | 2012-05-22 | Samsung Electronics Co., Ltd. | Memory module, a memory system including a memory controller and a memory module and methods thereof |
KR100759780B1 (ko) * | 2006-09-05 | 2007-09-20 | 삼성전자주식회사 | 반도체 메모리 장치 및 반도체 메모리 장치의 데이터 입출력 방법 |
WO2008090409A2 (en) * | 2006-10-04 | 2008-07-31 | Marvell Technology Japan Y.K. | Flash memory control interface |
US8028186B2 (en) | 2006-10-23 | 2011-09-27 | Violin Memory, Inc. | Skew management in an interconnection system |
TW200901042A (en) * | 2007-06-23 | 2009-01-01 | Jmicron Technology Corp | Storage device and circuit element switching method thereof |
JP2009023118A (ja) * | 2007-07-17 | 2009-02-05 | Canon Inc | インクジェット記録ヘッドのワイピング方法 |
US8397011B2 (en) * | 2007-10-05 | 2013-03-12 | Joseph Ashwood | Scalable mass data storage device |
JP5253901B2 (ja) * | 2008-06-20 | 2013-07-31 | 株式会社東芝 | メモリシステム |
US20100017569A1 (en) * | 2008-07-16 | 2010-01-21 | Agere Systems Inc. | Pcb including multiple chips sharing an off-chip memory, a method of accessing off-chip memory and a mcm utilizing fewer off-chip memories than chips |
US7957173B2 (en) * | 2008-10-14 | 2011-06-07 | Mosaid Technologies Incorporated | Composite memory having a bridging device for connecting discrete memory devices to a system |
US8134852B2 (en) * | 2008-10-14 | 2012-03-13 | Mosaid Technologies Incorporated | Bridge device architecture for connecting discrete memory devices to a system |
US8549209B2 (en) * | 2008-11-04 | 2013-10-01 | Mosaid Technologies Incorporated | Bridging device having a configurable virtual page size |
US20100115172A1 (en) * | 2008-11-04 | 2010-05-06 | Mosaid Technologies Incorporated | Bridge device having a virtual page buffer |
KR101375466B1 (ko) * | 2009-01-12 | 2014-03-18 | 램버스 인코포레이티드 | 다중 전력 모드를 갖는 메조크로노스 시그널링 시스템 |
KR101094402B1 (ko) | 2009-12-29 | 2011-12-15 | 주식회사 하이닉스반도체 | 반도체 장치 및 반도체 장치를 포함하는 반도체 시스템 |
US9542343B2 (en) | 2012-11-29 | 2017-01-10 | Samsung Electronics Co., Ltd. | Memory modules with reduced rank loading and memory systems including same |
CN104239252A (zh) * | 2013-06-21 | 2014-12-24 | 华为技术有限公司 | 数据存储系统的数据传输方法、装置及系统 |
JP5782556B2 (ja) * | 2014-10-15 | 2015-09-24 | 株式会社日立製作所 | 不揮発半導体記憶システム |
JP5968508B2 (ja) * | 2015-07-17 | 2016-08-10 | 株式会社日立製作所 | 不揮発半導体記憶システム |
US10705868B2 (en) * | 2017-08-07 | 2020-07-07 | Modelop, Inc. | Dynamically configurable microservice model for data analysis using sensors |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01296327A (ja) * | 1988-05-25 | 1989-11-29 | Seiko Instr Inc | コンピュータのバスライン |
JPH0215356A (ja) * | 1988-07-04 | 1990-01-19 | Hitachi Ltd | 信号バス分離構造 |
JPH086849A (ja) * | 1994-06-16 | 1996-01-12 | Kofu Nippon Denki Kk | 半導体記憶装置 |
JPH1055668A (ja) * | 1996-08-13 | 1998-02-24 | Fujitsu Ltd | 半導体集積回路、半導体集積回路モジュール、および、半導体集積回路システム |
JPH10124210A (ja) * | 1996-10-25 | 1998-05-15 | Hitachi Ltd | バスシステム及び回路基板 |
JPH10228449A (ja) * | 1996-09-17 | 1998-08-25 | Fujitsu Ltd | 半導体装置システム及び半導体装置 |
WO1999046685A1 (fr) * | 1998-03-12 | 1999-09-16 | Matsushita Electric Industrial Co., Ltd. | Selecteur de bus et systeme de circuit integre |
Family Cites Families (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US2002A (en) * | 1841-03-12 | Tor and planter for plowing | ||
US2000A (en) * | 1841-03-12 | Improvement in the manufacture of starch | ||
US2001A (en) * | 1841-03-12 | Sawmill | ||
US3593167A (en) * | 1969-01-28 | 1971-07-13 | Honeywell Inc | Synchronous read clock apparatus |
JPS6289149A (ja) * | 1985-10-15 | 1987-04-23 | Agency Of Ind Science & Technol | 多ポ−トメモリシステム |
WO1994003901A1 (en) * | 1992-08-10 | 1994-02-17 | Monolithic System Technology, Inc. | Fault-tolerant, high-speed bus system and bus interface for wafer-scale integration |
US5893919A (en) * | 1996-09-27 | 1999-04-13 | Storage Computer Corporation | Apparatus and method for storing data with selectable data protection using mirroring and selectable parity inhibition |
JP2935694B2 (ja) | 1997-04-25 | 1999-08-16 | 松下電器産業株式会社 | 半導体集積回路およびシステム、並びにクロック信号とデータ信号との間のスキューを低減する方法 |
US6011710A (en) * | 1997-10-30 | 2000-01-04 | Hewlett-Packard Company | Capacitance reducing memory system, device and method |
US7007130B1 (en) * | 1998-02-13 | 2006-02-28 | Intel Corporation | Memory system including a memory module having a memory module controller interfacing between a system memory controller and memory devices of the memory module |
US6349051B1 (en) * | 1998-01-29 | 2002-02-19 | Micron Technology, Inc. | High speed data bus |
US6742098B1 (en) * | 2000-10-03 | 2004-05-25 | Intel Corporation | Dual-port buffer-to-memory interface |
JP3644265B2 (ja) | 1998-08-04 | 2005-04-27 | 株式会社日立製作所 | メモリサブシステム |
DE19838813A1 (de) | 1998-08-26 | 2000-03-02 | Siemens Ag | Speichersystem |
US6628605B1 (en) * | 1999-07-21 | 2003-09-30 | Conexant Systems, Inc. | Method and apparatus for efficiently transmitting multiple data signals |
US6708248B1 (en) * | 1999-07-23 | 2004-03-16 | Rambus Inc. | Memory system with channel multiplexing of multiple memory devices |
JP3892655B2 (ja) | 1999-09-17 | 2007-03-14 | 株式会社東芝 | 半導体集積回路装置 |
US6530033B1 (en) * | 1999-10-28 | 2003-03-04 | Hewlett-Packard Company | Radial arm memory bus for a high availability computer system |
JP2001169068A (ja) * | 1999-12-14 | 2001-06-22 | Ricoh Co Ltd | 画像処理装置、画像処理方法およびその方法をコンピュータに実行させるプログラムを記録したコンピュータ読み取り可能な記録媒体 |
US6502161B1 (en) * | 2000-01-05 | 2002-12-31 | Rambus Inc. | Memory system including a point-to-point linked memory subsystem |
JP3821678B2 (ja) * | 2001-09-06 | 2006-09-13 | エルピーダメモリ株式会社 | メモリ装置 |
DE60210170T2 (de) * | 2002-07-15 | 2006-11-02 | Infineon Technologies Ag | Speichersystem |
US7206891B2 (en) * | 2002-09-26 | 2007-04-17 | Lsi Logic Corporation | Multi-port memory controller having independent ECC encoders |
US7198197B2 (en) * | 2002-11-05 | 2007-04-03 | Rambus, Inc. | Method and apparatus for data acquisition |
-
2000
- 2000-06-21 JP JP2000186234A patent/JP2002007201A/ja active Pending
-
2001
- 2001-06-20 WO PCT/JP2001/005270 patent/WO2001098880A1/ja active IP Right Grant
- 2001-06-20 CN CN2009101371262A patent/CN101561707B/zh not_active Expired - Fee Related
- 2001-06-20 US US10/311,687 patent/US7366821B2/en not_active Expired - Lifetime
- 2001-06-20 TW TW090115035A patent/TW511085B/zh not_active IP Right Cessation
- 2001-06-20 CN CNB018114016A patent/CN100504727C/zh not_active Expired - Fee Related
- 2001-06-20 KR KR10-2002-7017505A patent/KR100512895B1/ko not_active IP Right Cessation
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01296327A (ja) * | 1988-05-25 | 1989-11-29 | Seiko Instr Inc | コンピュータのバスライン |
JPH0215356A (ja) * | 1988-07-04 | 1990-01-19 | Hitachi Ltd | 信号バス分離構造 |
JPH086849A (ja) * | 1994-06-16 | 1996-01-12 | Kofu Nippon Denki Kk | 半導体記憶装置 |
JPH1055668A (ja) * | 1996-08-13 | 1998-02-24 | Fujitsu Ltd | 半導体集積回路、半導体集積回路モジュール、および、半導体集積回路システム |
JPH10228449A (ja) * | 1996-09-17 | 1998-08-25 | Fujitsu Ltd | 半導体装置システム及び半導体装置 |
JPH10124210A (ja) * | 1996-10-25 | 1998-05-15 | Hitachi Ltd | バスシステム及び回路基板 |
WO1999046685A1 (fr) * | 1998-03-12 | 1999-09-16 | Matsushita Electric Industrial Co., Ltd. | Selecteur de bus et systeme de circuit integre |
Non-Patent Citations (1)
Title |
---|
NORIHIKO TADANO, DIRECT RAMBUS: "1999 nen no pasokon nushi kioku wo nerau (Vol. 2) protocol ya denryoku seigyo kinou wo saiteki-ka", NIKKEI ELECTRONICS, no. 710, 23 February 1998 (1998-02-23), pages 163 - 176, XP002945862 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100461085C (zh) * | 2002-09-18 | 2009-02-11 | 株式会社日立制作所 | 存储系统及其控制方法 |
Also Published As
Publication number | Publication date |
---|---|
CN1437718A (zh) | 2003-08-20 |
CN101561707A (zh) | 2009-10-21 |
US20030163606A1 (en) | 2003-08-28 |
JP2002007201A (ja) | 2002-01-11 |
TW511085B (en) | 2002-11-21 |
KR100512895B1 (ko) | 2005-09-07 |
CN101561707B (zh) | 2011-09-14 |
US7366821B2 (en) | 2008-04-29 |
CN100504727C (zh) | 2009-06-24 |
KR20030012893A (ko) | 2003-02-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2001098880A1 (fr) | Systeme de memoire rapide | |
TWI447591B (zh) | 減少串列化器解串列化器鏈路中之潛時之技術 | |
US6625675B2 (en) | Processor for determining physical lane skew order | |
US8050332B2 (en) | System and method for selectively performing single-ended and differential signaling | |
US8700818B2 (en) | Packet based ID generation for serially interconnected devices | |
US7979611B2 (en) | Multi-protocol serial interface apparatus and system-on-chip apparatus including the same | |
EP1958404B1 (en) | Alignment and deskew for multiple lanes of serial interconnect | |
JP2009510657A5 (ja) | ||
JP4263105B2 (ja) | 高速の構成可能なトランシーバアーキテクチャ | |
US6249875B1 (en) | Interface circuit using plurality of synchronizers for synchronizing respective control signals over a multi-clock environment | |
JP7266698B2 (ja) | 信号処理チップ、及び信号処理システム | |
US20230117385A1 (en) | Low latency retimer and low latency control method | |
JP3087258B2 (ja) | コンピュータ・プロセッサのネットワーク及びデータ半同期伝送方法 | |
JP3989376B2 (ja) | 通信システム | |
CN100550828C (zh) | 以太网数据包与多个dsp串口数据转发的方法和系统 | |
US7380084B2 (en) | Dynamic detection of block boundaries on memory reads | |
US20140244868A1 (en) | Integrated circuit devices, systems and methods having automatic configurable mapping of input and/or output data connections | |
US7213090B2 (en) | Data transfer apparatus for serial data transfer in system LSI | |
US6701396B2 (en) | Data burst transfer circuit, parallel-serial and serial-parallel conversion circuits, and an oscillation circuit | |
US7515667B2 (en) | Method and apparatus for reducing synchronizer shadow | |
US20050152205A1 (en) | Semiconductor memory | |
WO2021012767A1 (zh) | 用于存储器控制器与存储器设备互连的总线 | |
JP5902402B2 (ja) | データ出力調整装置、データ出力調整方法、rgmiiネットワークシステム、及び、rgmiiネットワーク通信路切替方法 | |
US7269681B1 (en) | Arrangement for receiving and transmitting PCI-X data according to selected data modes | |
KR101346293B1 (ko) | 디시리얼라이저 및 데이터 복원 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 018114016 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020027017505 Country of ref document: KR |
|
WWP | Wipo information: published in national office |
Ref document number: 1020027017505 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10311687 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase | ||
WWG | Wipo information: grant in national office |
Ref document number: 1020027017505 Country of ref document: KR |