JPH02272648A - File rearrangement system - Google Patents

Abstract

PURPOSE:To improve the performance of a computer system by dynamically copying rearranged files. CONSTITUTION:A file rearrangement information input part 4 inputs and analyzes file rearrangement information and starts a rearranged file copying part 5 and the copying part 5 copies a rearranged file 2 based upon the file rearrangement information analyzed by the input part 4. A rearranged file access control part 6 executes the file access control processing of a file access request to the file 2 being copied by the copying part 5 and a file information updating part 7 updates file information relating to the rearranged file 2 after ending the copying of the rearranged copy 2 by the copying part 5. Consequently, file rearrangement can be executed without ending an on-line job and the performance of the computer system 10 can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a file relocation method, and particularly to a file relocation method in a file management method of a computer system.

[Conventional technology]

In computer system file management methods, computer system performance analysis (disk volume 10
(Input/output) If it is found that 10 is concentrated on a specific disk volume (load situation analysis, etc.), some files in that disk volume are relocated to other disk volumes (file relocation) A file relocation method is used to relocate files (the target file for file relocation is called a relocation file).

Conventionally, this type of file relocation method temporarily stops using the relocation file (pre-relocation file) before file relocation when copying the relocation file, and then restarts the file after copying the relocation file. The relocation file (post-relocation file) was being reused after placement.

[Problem to be solved by the invention]

In the conventional file relocation method described above, it is necessary to temporarily stop using the relocation file when copying the relocation file. ) is an online job, it is necessary to terminate the online zip itself in order to relocate the files, which has the disadvantage of causing a decrease in computer system performance.

In view of the above-mentioned points, an object of the present invention is to dynamically perform a file copy of a relocated file (a file copy of a relocated file can be performed without temporarily discontinuing the use of the relocated file). An object of the present invention is to provide a file relocation method that can perform file relocation without terminating an online job that is a job that uses relocation files, and can improve the performance of a computer system.

[Failure to solve the problem]

The file relocation method of the present invention includes a file relocation information input section that inputs file relocation information, analyzes it, and starts a relocation file copy section, and a file relocation method analyzed by the file relocation information input section. The relocated file copying unit performs a file copy of the relocated file based on the information, and the relocated file access management performs file access management processing in response to a file access request to the relocated file during file copying by the relocated file copying unit. and a file information update unit that updates file information regarding the relocated file after the relocated file copy unit finishes copying the relocated file.

[Effect]

In the file relocation method of the present invention, the file relocation information input section inputs and analyzes file relocation information and starts the relocation file copy section, and the relocation file copy section analyzes the file relocation information human power section. The relocated file access management unit performs file access management processing in response to a file access request to the relocated file during file copying by the relocated file copy unit, and The file information update unit updates the file information regarding the relocated file after the file copy of the relocated file by part of the relocated file copy is completed.

〔Example〕

Next, the present invention will be explained with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the file relocation method of the present invention. The file relocation method of this embodiment is based on terminal l, file 2 before relocation, and rewrite gll.
It is configured to include a file 3 and a computer system IO.

The computer system 10 includes a file relocation information input unit 4 that inputs file relocation information, which is information regarding the pre-relocation file 2 and the post-relocation file 3, from the terminal 1, analyzes it, and starts the relocation file copy unit 5. and a relocation file copy unit 5 that copies the contents (data) of the pre-relocation file 2 to the post-relocation file 3;
Performs file access management processing in response to a file access request to the relocated file from the relocated file usage job 9 during file copying of the relocated file (a concept that includes the pre-relocated file 2 and the post-relocated file 3) A relocation file access management unit 6, a file information update unit 7 that updates file information 8 regarding the relocation file after the file copy of the relocation file is completed, and a pre-relocation file 2 and a post-relocation file 3.
File information about Il! 8, and a relocation file use job 9 that uses the pre-relocation file 2 and the post-relocation file 3.

FIG. 2 is a flowchart showing the processing (file access management processing) of the relocated file access management section 6. The processing of the relocated file access management unit 6 consists of a read determination step 21, an at-trace determination step 22, an output processing step 23, a both output processing step 24, and an input processing step 25.

Next, the operation of the file relocation method of this embodiment configured as described above will be explained.

The user of the computer system 10 analyzes the status of the disk volume ■0 load and relocates files in the disk volume with a high load to other disk volumes to distribute the IO load. This file is determined as the pre-relocation file 2.

In addition, the user determines the file to be created in the other disk volume as the file 3 after relocation in order to distribute the IO load, and in advance,
Reserve a file area.

When the pre-relocation file 2 and post-relocation file 3 are determined as described above, the user sends the file relocation information, which is information regarding the pre-relocation file 2 and the post-relocation file 3, to the computer from the terminal 1. system 10
[Input to the information input section 4]

The file relocation information input section 4 analyzes the input file relocation information and starts the relocation file copy section 5 if the file relocation information is valid.

The relocated file copy unit 5 sequentially copies the contents (data) in the pre-relocated file 2 to the post-relocated file 3 from the beginning.

If a file access request to the relocated file is issued from the relocated file usage job 9 while the relocated file copy unit 5 is performing the above-mentioned file copy,
The relocation file access management unit 6 performs file access management processing as shown below (see Figure 2). File relocation information regarding file 3 has been obtained).

First, a read determination is made to determine whether or not the file access request is a request to read data in a relocated file (step 21).

If the file access request is a continuation request in this read determination, an input process is performed to input data for the read request from the pre-relocation file 2 (step 25).

If the file access request is not a read request in the read judgment in step 21, it is a write request for data in the relocation file, so the address on the relocation file of the data that is the target of the write request is data that has already been copied to the file. Address determination is performed to determine whether the address is (step 22).

In this address determination, if the address on the relocation file of the data that is the target of the write request is the address of data that has already been copied, the address of the target of the write request will be added to both the file 2 before relocation and the file 3 after relocation. Perform re-output processing to output data (step 24
).

If the address on the relocation file of the data that is the target of the write request in step 22 is not the address of the file-copied data, output processing is performed to output the data only to the pre-relocation file 2 (step 23 ).

The above-described file access management processing by the relocation file access management unit 6 allows file access from the relocation file use job 9 even during file copying of the relocation file, even if the computer system 10 goes down. The contents of the pre-relocation file 2 can now be guaranteed.

When the file copy of the relocated file is completed as described above, the file information update unit 7 is started (the file information update unit 7 receives the file 2 before relocation and the file 3 after relocation from the file relocation information input unit 4). ).

The file information update unit 7 rewrites the content of the file information 8 regarding the file 2 before relocation to the content regarding the file 3 after relocation.

As a result, all subsequent file accesses by the relocation file use job 9 to the relocation file will be executed to the post-relocation file 3.

〔Effect of the invention〕

As explained above, the present invention dynamically relocates files (
By performing the D file copy, file rearrangement can be performed without terminating the online job, which is a job using M-arranged files, and the performance of the computer system can be improved.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, and FIG. 2 is a flow chart showing the processing of the relocation file access management section in FIG. In the figure, ■... Terminal, 2... File before relocation, 3... File after relocation, 4... File relocation information human resource department, 5... Relocation file copy unit, 6. ... Relocation file access management unit, 7... File information update unit, - File information, - Relocation file usage job, - Combiator system.

Claims (1)

[Claims] A file relocation information input section that inputs and analyzes file relocation information and starts a relocation file copy section; a relocation file copying unit that performs a file copy of a relocation file; a relocation file access management unit that performs file access management processing in response to a file access request to the relocation file during file copying by the relocation file copying unit; A file relocation method comprising: a file information update unit that updates file information regarding the relocated file after the relocation file copy unit finishes copying the relocated file.