JPS62121532A - Data retrieving method - Google Patents

Abstract

PURPOSE:To extract data in sorting order at a high speed by generating a relative address table of all page numbers of data to be sorted in a main storage device or external storage device, and sorting and rearranging the data according to the table. CONSTITUTION:Data is represented by page numbers of a disk file and in-page relative addresses and the sorting order is expressed by their arrangement order, so sorting results are stored with small storage capacity. Then, the stored sorting results become more than those in tree structure of each page and the number of times of input to and output from the buffer 8 and external storage device 4 for knowing the sorting order decreases. When one data is outputted, a reference to the sorting table 7 and a reference to data based upon the sorting table are both necessary, but reference records in the sorting table 7 are already present on the buffer frequently, so the input/output operation of the external storage device 7 is performed only once. Thus, data are processed at a high speed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a data processing method for extracting and processing data from a file, and particularly to a data retrieval method suitable for extracting a large amount of data in a specified order.

[Background of the invention]

Conventionally, data is expressed in a tree structure consisting of a certain item that constitutes it as a key and its order relationship.Each node in the tree has a key value, a pointer to a lower/upper level node, a pointer to data, etc. It consists of The input/output time between the main storage device and the external storage device accounts for a large proportion of the data processing time, and in order to process data at high speed, it is important to reduce the number of input/output operations. In tree-structured data, when data is retrieved into keys, the next data is obtained by tracing the branches of the tree, but it is better to read a large amount of tree-structured information into main memory at once, which reduces the number of input/outputs. It is possible to increase the speed by reducing the number of times. Therefore, the problem is how to express the information necessary for extracting data to keys with a small amount of information.

For information related to data retrieval, please refer to Japanese Unexamined Patent Publication No. 58-3.
No. 032, JP-A-58-97742, JP-A-58-1
No. 15543 etc. are mentioned.

[Purpose of the invention]

The purpose of the present invention is to express sorting results with a small amount of information,
An object of the present invention is to provide a data retrieval method that allows data to be sorted and retrieved at high speed.

[Summary of the invention]

The present invention includes a main storage device storing a sorting processing section, a CPU
, in a system including an external storage device,
A table is provided in the main storage device or the external storage device, and the data is outputted based on the table when data is output.

[Embodiments of the invention]

An embodiment of the method of the present invention will be described below with reference to FIGS. 1 to 3. FIG. 1 is a system configuration diagram in this embodiment, FIG. 2 is an explanatory diagram of a sorting table according to the present invention, and FIG. 3 is a flowchart showing the processing procedure of the sort processing section 9 shown in FIG. 1.

As shown in Fig. 1, this system consists of an input device 1 for inputting data and various commands, an output device 2 for displaying a CRT or the like and displaying a hard copy, and forming the core of the system and performing various processing. CPU 3, main storage device 4.
It consists of a magnetic disk 5 which is an external storage device. A file on the magnetic disk 5 stores data 6 and a sorting table 7 representing the sorting results. The main storage device 4 includes a buffer 8 that temporarily stores data and a sort processing section 9 that stores a program for executing sort processing. The sort processing section 9 includes a sort execution section 9-1 and a sort result output section 9-2.

Sort table 7, as shown in Figure 2, shows the page number (page number) in which the corresponding data exists in the disk file.
It consists of two items: 7-1 and the data address (intra-page relative address) 7-2 when counted from the top of the page.

In order to extract the data into sort categories, the processing procedure shown in FIG. 3 is used. When a request comes to sort the data,
The sort execution unit 9-1 obtains the page number 7-1 and position within the page 7-2 in the disk 5 in which the data is stored for all the data to be sorted, and displays them in the table (7-0
) and put all of them together to obtain sort table 7. To refer to the contents of data 6 and sort table 7 on disk 5,
A magnetic disk 5 and a buffer 8 containing data to be referenced
A page portion, which is an input/output unit, is loaded from the magnetic disk 5 to the buffer 8 and referenced.

The sort execution unit 9-1 can load the page into the buffer 8 using the page number 7-1 of 7-0, and refer to the data using the relative address 7-2. By referring to the data in this way, the sort execution unit 9
-1 performs sorting and rearranges records 7-0 corresponding to the corresponding data. In this way, the sort table 7 is created. When outputting data in sort 1-order, the sort result output unit 9-2 loads each page of the sort table 7 into the buffer 8 in order from the first page of the sort table 7, and outputs the page numbers in order from the first page of the sort table 7. Data can be loaded into the buffer based on 7-1 and the relative address 7-2, and the data can be output to the sort by referring to it.

In this manner, in this embodiment, data is represented by a page number of a disk file and a relative address within the page, and sorting is represented by the order of sorting, so that sorting results can be stored with a small storage capacity. As a result, more sort results are stored per page than in the case of a tree structure, and the number of inputs and outputs between the buffer and the external storage device required to know the sorts is reduced. To output 1 data, it is necessary to refer to sort table 7 twice, and to refer to data based on the table to sort 1, but in most cases, the reference record of sort table 7 is already on the buffer. Therefore, the number of inputs and outputs to and from the external storage device is only one, and data can be processed at high speed.

In this embodiment, the case where the sort table 7 is stored on the magnetic disk is explained, but if the number of data is small or
If the buffer capacity is sufficiently large, it can also be located on the main memory.

〔Effect of the invention〕

According to the present invention, a sort table whose item is the storage location of data in an external storage device is used, and data corresponding to the records of the sort table is output, so the sort results can be expressed in a small storage area, and can be expressed in one page. Many sort results can be stored inside. As a result, the number of inputs and outputs between the buffer and the external storage device for referring to the sort table can be reduced.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of a system for realizing the present invention, and FIG. 2 is a sorting table 7 obtained by the present invention.
FIG. 3 is a flowchart showing an embodiment of the present invention. 1...Input device, 2...Output device, 3...CPU,
4. Main storage device, 5.. Magnetic disk, 6.. Data, 7.. Sort table, 8.. Buffer, 9.. Sort processing unit. Agent Patent Attorney Tadashi Akimoto Figure 1 Figure 2

Claims (1)

[Claims] In a system equipped with a central processing unit that executes sort processing, a main storage device that stores a program for sort processing, and an external storage device that stores data, the relative page number of all data to be sorted is calculated. A data retrieval method characterized in that a table of addresses is created in the main storage device or the external storage device, and records corresponding to corresponding data are rearranged by referring to the table and performing sorting.