JPS63291127A - Data processor - Google Patents

Abstract

PURPOSE:To speed up a matching processing by executing the comparison in parallel under the comparing conditions set to respective elements of the list data of a table format. CONSTITUTION:When a matching processing is executed, matching data are set to a comparing register 4 in a list data processing part 6, further, when comparing conditions are set to a matching register 5, the data of the comparing register 4 are compared under the comparing conditions of the matching register 5 to the data of respective element registers 11 by a comparing means 13 in respective subarithmetic units 14 and said elements are searched. Thus the high speed of the matching processing can be realized.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a data processing device mainly intended for use in the field of artificial intelligence.

BACKGROUND OF THE INVENTION In recent years, the field of artificial intelligence has been actively researched as one of computer applications. In this field, it is necessary to process structured data, and therefore LISP, a language that can handle structured data, is widely used. However, since it is inefficient to execute the LISP language on a general-purpose computer, special-purpose machines with various improvements have been developed.

These dedicated machines were improved mainly from a linguistic perspective, and the typical improvements made are shown below.

+l) Primitive functions such as CAR and CD R are executed at the microprogram level.

(2) TAG for handling generic data types
data format.

(3) A hardware air control stand is provided to speed up stack processing.

(References rLIsP Machine” Information Processing Vol. 23
8 pp752-772) However, although improvements have been made to the execution system of the language as described above, the representation of structure data inside a computer basically uses pointers to represent the order of elements and the method of connection. (hereinafter referred to as the list). Since this method involves the operation of sequentially following the pointer to all list originals, it is essentially inefficient for performing the following operations that frequently appear in application programs in the field of artificial intelligence.

*Pattern matching It is inefficient because it involves list decomposition.

* Access to arbitrary elements, decomposition of lists, and replacement lists for specific elements are inefficient.

In order to solve the above problems, proposals have been made to change the method of expressing list data.

-a, the binary tree list starts from the starting point node and sequentially branches left and right, with each branch ending at a leaf node. There are two types of leaf nodes: atom nodes and NIL nodes. Nodes that are not leaf nodes are list nodes that indicate that the branch is continuing. This list node is used to indirectly indicate the position of the leaf node.

In the pointer representation, this structural representation is expressed as is by cells in which all nodes are connected by addresses. As a result, access to each element always requires repeated indirect references to the address.

However, if the positions of leaf nodes can be directly determined, there is no need to have list node information. Therefore, in a table that sequentially arranges the leaf position information and the leaf's own information,
Equivalent list data can be expressed. As a method for expressing leaf node positions, a one-dimensional Bertol representation has been proposed in which numbers are sequentially assigned in the CDR direction and items are sequentially assigned in the CAR direction. Therefore, the list data is expressed as a set of data that includes a vector indicating leaf position information and information about the leaf itself. With this representation method, there is no need to limit the interpretation of a list to a binary tree list, as is the case with pointer representations.

FIG. 2 shows an example of a tabular representation of list data. This shows the graphical representation (11) and tabular representation (2) of list data that becomes (A (B (C) ) D) when expressed in 8 formulas. In the diagrammatic representation, circles indicate list data. It represents a node, and the boxed area indicates a leaf node.The number string appended to each node indicates the node position expressed according to the method described above.Extract this leaf part. The table representation (2) is a table representation (2) in which the ADDRESS section contains the node position vector and the VALUE section contains the leaf elements. This makes it possible to process list data in parallel for each element without passing through pointers, and operations such as pattern matching, accessing arbitrary elements, disassembling lists, and replacing specific elements can be performed. It can be done quickly.

An example of a conventional data processing device based on this tabular representation will be described below with reference to the drawings.

FIG. 3 shows a conventional processing device. In FIG. 3, 31 is a memory device for storing tabular list data, 32 is a list-data processing device comprising a plurality of processing units 33 that process each element of the tabular data, 34 is an atom data processing device, and 535 36 is a data comparison device that performs matching on the elements of each processing unit 33 in the list data processing device 32, and 36 is each element of tabular data from the memory device 31 to each processing unit 33 in the list data processing device 32. allocation and each processing unit 3
This is a transfer device that transfers the elements in 3 to the data comparison device 35.

The operation of the data processing apparatus configured as described above will be described below.

First, normal numerical data, character data, etc. are processed in the atom data processing device 34.

The list data is transferred from the memory device 31 to each processing unit 33 in the list data processing device 32 for each element by the device 3.
6 and processed in parallel by each element by each processing unit 33. By processing list data in parallel for each element separately from atom data, we aim to efficiently process list data and simplify the configuration of the processing system.

Further, when comparing data in each processing unit 33 by matching or the like, the data is transferred to the data comparison device 35 by the transfer device 36 and compared.

Problems to be Solved by the Invention However, with the above configuration, when performing comparisons such as matching on the elements of each processing unit in the list data processing device, data transfer from each processing unit to the data comparison device is difficult. 2. The comparison process takes time because it occurs frequently and the comparison process is performed by software. This method has a problem in that a bottleneck occurs in parallel processing of list data.

In view of the above-mentioned problems, the present invention provides a data processing device that reduces the amount of data transferred in comparison processing of list data in a tabular format and speeds up the comparison processing.

Means for Solving the Problems In order to solve the above problems, the data processing device of the present invention includes a memory device that stores list data as tabular data in which the position of each element is expressed as a vector; A list data processing section consisting of a plurality of list registers and a list calculation means for storing identification information of list data, a comparison register for storing comparison data, and a matching register for storing comparison conditions, and an atom consisting of a plurality of registers and a calculation means. It has a main processing unit consisting of a data processing section, a plurality of element registers for storing each element of list data in a tabular format, element calculation means, and comparison means, and is capable of processing list data under the control of the main processing unit. a plurality of sub-processing units that process each element in parallel; a transfer unit that sequentially allocates elements of the tabular data stored in the memory device to each sub-processing unit; It is equipped with an address comparison device that compares in parallel.

According to the above-described structure, the present invention simultaneously compares the data stored in the comparison register for each element of tabular list data under the comparison conditions stored in the matching register. Since data transfer can be avoided, and comparison conditions can be set arbitrarily, data matching processing can be performed at high speed.

Embodiment Hereinafter, a data processing apparatus according to an embodiment of the present invention will be described from the beginning with reference to the drawings. FIG. 1 shows the configuration of a data processing device in an embodiment of the present invention. In FIG. 1, l is a main memory device, 2 is a list register, and 3 is a main memory device.
4 is a list calculation means, 4 is a comparison register, and 5 is a matching register, and the list data processing unit 6 is a general term for a plurality of list registers, list calculation means, comparison register, and matching register. 7 is a register, 8 is an arithmetic means, and the atom data processing section 9 is a general term for a plurality of registers and arithmetic means. Furthermore, the main processing unit 8 is a general term for the list data processing section and the atom data processing section. 11
12 is an element register, 12 is an element calculation means, 13 is a comparison means, and the sub-operation unit 14 is a general term for the element register, element calculation means, and comparison means. 15 is a transfer device.

The data processing apparatus configured as above will be explained below using FIG. 1 and FIG. 2. Here, FIG. 2 shows an example of a tabular representation of list data.

This is expressed in equation 8 (A (B (C,) )
D) shows the diagrammatic representation (11) and tabular representation (2) of the list data. In the diagrammatic representation, circles represent list nodes, and squares represent leaf nodes. In addition, the number string added above each node indicates the node position expressed according to the method described above.The table format representation (2) extracts this leaf part and represents it in the form of a table. It consists of a table with node position vectors in the ADDRESS section and leaf elements in the VALUE section.

First, a reference to atom data is stored in the VALUE section of each element of list data.

The processing of numerical and character data, which are their actual values, is
Register 7 and calculation means 8 in atom data processing section 9
This is done using These atomic data are stored in memory device l. Next, the list data is processed by the list data processing section 6 and the sub-processing unit 14 in the main processing unit 10. For one list register 2 of the list data processing unit, the element register 11 in each sub-processing unit
One of them corresponds. That is, identification information of list data is stored in the list register 2, and each element of the list data is stored in a plurality of corresponding element registers, representing one list data as a whole. When storing or processing list data in the memory device 1, the identification information of the list data is first transferred to the list register 2 in the list data processing section 6, and at the same time, each element of the list data is transferred to each subsystem by the transfer device 15. They are separately transferred to the element registers 11 of the arithmetic unit 14. When the operation is performed on the list register 2, the necessary operation is performed on the corresponding element register 11 at the same time.

When performing matching processing, matching data is set in the comparison register 4 in the list data processing section 6, and furthermore, a comparison condition is set in the matching register 5. The data in the comparison register 4 is compared with the data in the element register 11 under the comparison conditions in the matching register 5, and the corresponding element is searched.

Condition settings to the matching register are '>', '<,' for the comparison data to be set in the comparison register.
In addition to the relationships >=', '=<', and '=', it is also possible to set a wild card that applies a comparison condition only to a specified portion of comparison data.

As described above, the present embodiment includes a comparison register for setting comparison data, a matching register for setting comparison conditions, and comparison means for performing parallel comparison for each element of each sub-processing unit. However, by performing parallel comparisons based on the comparison conditions set for each element of the tabular list data, it is possible to speed up the matching process.

Effects of the Invention As described above, the present invention includes a memory device that stores list data as tabular data in which the position of each element is expressed as a vector, and a plurality of list registers that store identification information of the tabular list data. A main processing unit consisting of a list data processing unit consisting of a list calculation means, a comparison register for storing comparison data, and a matching register for storing comparison conditions, and an atom data processing unit consisting of a plurality of registers and calculation means; a plurality of sub-processing units that are configured of a plurality of element registers for storing each element of the list data in the format, an element calculation means, and a comparison means, and that process each element of the list data in parallel under the control of the main processing unit; and a transfer device that sequentially allocates elements of the tabular data stored in the memory device to each of the sub-processing units, and transfers the elements of the tabular list data in parallel based on comparison conditions set for each element of the tabular list data. The matching process can be sped up by comparing the results.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of a data processing device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram of a tabular representation of list data according to an embodiment of the present invention, and FIG. 3 is a configuration diagram of a conventional data processing device. It is a diagram. 1...Main memory device, 2...List register, 3...List calculation means, 4...
・Comparison register, 5...Matching register,
6...List data processing unit, 7...Register, 8...Calculating means, 9...Atom data processing unit, 10... Main processing unit, 11.
...Element register, 12 ...Element operation means, 13 ...Comparison means, 14 ...Sub-operation device, 15 ...Transfer device, 31...
Memory device, 32...List data processing device,
33... Processing unit, 34... Atom data processing device, 35... Data comparison device,
36...Transfer device. Name of agent: Patent attorney Toshio Nakao
Figure 2 of Ng Castle (A (B (c)) D)

Claims (1)

[Claims] A memory device that stores list data as tabular data in which the position of each element is expressed as a vector, a plurality of list registers that store identification information of the tabular list data, a list calculation means, and a comparison register that stores comparison data. a main processing unit consisting of a list data processing section consisting of a matching register that stores comparison conditions; and an atom data processing section consisting of a plurality of registers and calculation means; a plurality of sub-processing units that are composed of element registers, element calculation means, and comparison means and process each element of list data in parallel under the control of the main processing unit; and the memory device for each of the sub-processing units. 1. A data processing device comprising a transfer device that sequentially allocates elements of tabular data stored in a data processing device.