JPH0216651A - Disk cache control system - Google Patents

Abstract

PURPOSE:To speed up the disk access and sort processing of the title system by using the entire area of a cache block as a disk cache area at normal processing time and, when a process which uses a large area of a main memory device occurs, releasing the blocks of the number required for the process. CONSTITUTION:In a disk cache area 5, the entire area is allocated to cache blocks after system rising so that the disk access to a disk device 3 can be speeded up. When a sorting process is produced from a central processor 1 in such state, the cache blocks of the area required for the sorting process of the area 7 for both disk cache and sorting are released and the area is allocated to the necessary sorting area. Thereafter, the device 1 performs the sorting process at high speed by using the sorting area in accordance with the program of a program area 4 and, upon completion of the sorting process, releases the sorting area to the original blocks so that the entire area 5 can be used as a cache again.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a disk cache control method for controlling a disk cache area on a main storage device that stores input/output data for a disk device in a data processing device.

[Conventional technology]

FIG. 6 is a block diagram of a data processing device that employs a conventional disk cache control method.

In the figure, 1 is a central processing unit that performs calculations and controls necessary for data processing, 2 is a main memory that stores programs and data necessary for the above calculations and controls, and 3 is a storage unit that stores data processed by the central processing unit 1. This is a disk device that supports The main storage device 2 has a program area 4 that stores the above programs, a disk cache area 5 that stores input/output data for the disk device 3, and a sort 914 area 10 that stores data related to sort processing. This data processing device utilizes the disk cache area 5 in order to input data from the disk device 3 at high speed when executing data processing.

FIG. 7 is for explaining the configuration of the disk cache area in FIG. 6. The disk cache area 5 is
It is composed of a plurality of cache blocks 8 of a certain size, and holds a copy of the contents stored in the data area 9 on the disk device 3.

Next, the operation will be explained. In the conventional data processing device as shown in FIG. 6, the disk cache area 5 on the main storage device 2 is used as follows.

The disk cache area 5 permanently occupies the main storage device 2 and has a plurality of cache blocks 8 as shown in FIG. These cache blocks 8 hold a copy of the contents of the data area 9 on the disk device 3. When the program in the program area 4 reads data from the disk device 3, if the data exists in the cache block 8, the data is read from the cache block 8, and if the data does not exist in any cache block 8, the data is newly cached. A copy of the contents of the data area 9 on the disk device 3 in which the data exists is allocated to the block 8, and then the data is read from the corresponding cache block 8.

[Problem to be solved by the invention]

In the conventional disk cache control method, as mentioned above, the disk cache area 5 occupies the main storage device 2 in a fixed manner. is main memory 2
A large amount of the above sort area 10 is required, and therefore the disk cache area 5 that occupies the limited main memory 2 is
The capacity of the main storage device 2 is also limited, and therefore a large amount of the main storage device 2 is required in order to access the disk device 3, sort processing, etc. at high speed.

This invention was made to solve the above-mentioned problems, and provides a disk cache control method that does not require a large amount of main storage and can perform disk device access, sort processing, etc. at high speed. The purpose is to

[Means to solve the problem]

In the disk cache control method according to the present invention, a disk cache area 5 is provided on the main storage device 2 and is composed of a plurality of cache blocks 8 and stores input/output data for the disk device 3, and the central processing unit 1 performs normal processing. At times, the entire cache block 8 is used as the disk cache area 5, and when a process that uses a large amount of space in the main storage device 2 occurs, the number of cache blocks 8 necessary for the process is released from the disk cache area 5. , these released cache blocks 8 are used for the processing, and when the processing is completed, the released cache blocks 8 are controlled to be returned to the original disk cache area 5. It is something to do.

[Effect]

The central processing unit 1 uses the cache block 8 during normal processing.
The entire area is used as the disk cache area 5 to speed up data transfer to the disk device 3. Furthermore, when a process that involves sorting a large amount of data occurs, the central processing unit 1 releases the number of cache blocks 8 necessary for the process from the disk cache area 5, and uses these released cache blocks 8. The processing described above that uses a large amount of the area of the main storage device 2 is executed. Further, when the central processing unit 1 completes the above processing, the central processing unit 1 returns the released cache block 8 to the original disk cache area 5.

[Embodiments of the invention]

FIG. 1 is a block diagram of a data processing apparatus employing a disk cache control method according to an embodiment of the present invention. In FIG. 1, components corresponding to those shown in FIG. 6 are designated by the same reference numerals, and their explanations will be omitted. The disk cache area 5 on the main storage device 2 in FIG.
Disk cache dedicated area 6 and disk cache/
It consists of a sorting area 7. The disk cache dedicated area 6 is a dedicated area for storing input/output data for the disk device 3. The disk cache/sort area 7 is an area that is used both as a disk cache area and as a sort area for sort processing, and is used as a disk cache area or as a sort ML area as required.

The operation of the disk cache area 5 will be explained based on the flowchart shown in FIG. After the system is started (step S1), the entire disk cache area 5 is allocated to cache blocks (step S2), and disk access to the disk unit W3 is performed at high speed (step S2).
Step S3).

Here, when the sorting process by the central processing unit 1 occurs (
In step S4), the cache blocks for the area necessary for sort processing in the disk cache/sort area 7 are released and allocated to the sort 9M area necessary for sort processing (step S5). Then, the central processing unit 1 performs the sorting process at high speed using the sorting area according to the program in the program area 4 (step s6), and when the sorting process is finished, the sorting area is restored to the released cache block (step s7), and then again. The entire disk cache area 5 is used as a disk cache.

Figure 3 shows the disk cache area 5 in this embodiment.
4 is a diagram showing a state in which the entire disk cache area is used as a disk cache, and FIG. 4 shows a state in which a part of the disk cache area is released to create a sort area l.

FIG. Normally, after the system is started up, the entire disk cache area 5, which is divided into a plurality of cache blocks 8 of a certain capacity as shown in FIG. 3, is used as a disk cache.
By allocating a copy of the contents of the data area 9 on the disk device M3 to the cache block 8, disk access is performed at high speed. When the sorting process occurs, as shown in the flowchart of FIG. 5, first of all, if there is a cache block 8 whose value remains changed after data allocation (step 5ll), that cache block 8 is The data of block 8 is written to the corresponding data area 9 on disk device 3 (step 512), and access synchronization between disk cache area 5 and disk device 3 is established. Next, calculate the memory size required for sorting based on the size and number of data to be sorted (step 513).
, the consecutive cache blocks 8 of that size on the main storage device 2 are invalidated (step 514) and released from the disk cache area 5. Thereafter, as shown in FIG. 4, the released cache block 8 is assigned 0 on the side of the soh fi Jt area 10 and is quickly sorted on the main storage device 2 (step 515).

After the sorting process is completed, the sort area 10 is released and restored to the disk cache area 5, and as shown in FIG. Make access.

In addition, in the above embodiment, the disk cache area is shown as a disk cache exclusive area and a disk cache/sorting area, but the disk cache area does not need to have a disk cache exclusive area, and the disk drive/7 system /The sorting area may be used for purposes other than sorting.

〔Effect of the invention〕

As described above, according to the present invention, during normal processing, the entire cache block is used as a disk cache area, and when a process that uses a large amount of main storage space occurs, the disk cache area is used to store the number of cache blocks necessary for the process. The system releases cache blocks, uses these freed cache blocks for processing, and controls the processing to return these released cache blocks to the original disk cache area once the processing is finished. , the capacity of the disk cache area can be changed dynamically, which eliminates the need for a large amount of main storage, and has the effect of speeding up disk device access, sorting, etc. . Further, according to the present invention, since the disk cache area is optimally changed according to the processing content, it is possible to effectively utilize a main storage device with a small capacity, and the cost of the device can also be reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a data processing device employing a disk cache control method according to an embodiment of the present invention, FIG. 2 is a flowchart for explaining the operation of the disk cache area in this embodiment, and FIG. Figure 3 shows a state in which the entire disk cache area is used as a disk cache in this embodiment, and Figure 4 shows a state in which a part of the disk cache area is released and used as a sort area. Figure 5 is a flowchart for explaining the operation of this embodiment, and Figure 6 is a flowchart for explaining the operation of this embodiment.
The figure is a block diagram showing the configuration of a data processing device employing a conventional disk cache control method, and FIG. 7 is a diagram for explaining the configuration of the disk cache area in FIG. 6. 1...Central processing unit, 2...Main storage device, 3...
Disk device, 5... Disk cache area, 8.
...cache block. Agent: Masu Oiwa A (and 2 others) Figure 1, Figure 2, Figure 3, Figure 4, 50 Figure 6, Figure 7, Procedural Amendment (Voluntary) Column of the drawing dated May 26, 2008 subject to amendment. Contents of the amendment The drawing, Figure 5, will be amended as shown in the attached sheet. Above 2. Name of the invention Disk cache control method 3. Relationship with the amendment person case

Claims (1)

[Claims] A central processing unit that performs calculations and controls necessary for data processing, a main storage device that stores programs and data necessary for the calculations and controls, and a disk device that stores data processed by the central processing unit. In the data processing device, a disk cache area is provided on the main storage device that is composed of a plurality of cache blocks and stores input/output data for the disk device, and the central processing unit stores all of the cache blocks during normal processing. is used as a disk cache area, and when a process occurs that uses a large amount of the main storage area, the number of cache blocks required for the process is released from the disk cache area, and these released cache blocks are used for that purpose. A disk cache control method characterized by controlling the cache blocks used for processing and returning these released cache blocks to the original disk cache area once the processing is completed.