JPH03126123A - Trouble recovering system for recording medium - Google Patents

Abstract

PURPOSE:To smoothly execute a normal input/output request during recovery by distributing and recording file contents to different mediums to quickly recover a system from disk trouble. CONSTITUTION:A means 11 is provided which distributes and records data to be recorded to different recording mediums 21 to 25 and successively reads out data from recording mediums 21 to 25 for the purpose of restoring data recorded on the recording medium, which is destroyed at the time of the occurrence of trouble, to write restored data in this recording medium and performs the normal read/write processing for a part of recording mediums, from which data is read out for the purpose of restoring data, and recording mediums other than the recording medium from which data is read out at present for the purpose of restoring data. Thus, recording mediums are quickly recovered from trouble, and the normal input/output request during recovery is smoothly executed.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention relates to a failure recovery method for a recording medium in a computer including a plurality of recording media and a central processing unit that writes/reads data to/from the plurality of recording media. [Background Art] In recent years, electronic computers have begun to write and save the same content on a plurality of recording media so that even if a failure occurs in a recording medium, it can be restored.

For example, when a failure occurs in one of the recording media, recovery from the failure is performed by reading normal data from a disk that serves as a restoration light and writing it to the disk that is the object of restoration.

Conventionally, in this type of restoration processing, multiplexing is performed for each recording medium, so that if a failure occurs in one disk, there is one disk for restoration light.

[Problems to be Solved by the Invention] The conventional restoration processing described above has a multiplexing configuration for each recording medium, so it is necessary to repeatedly access the same recording medium device when restoring.

This has the disadvantage that processing of normal input/output requests that occur during restoration is delayed, and restoration processing takes time.

[Means for Solving the Problems] It is an object of the present invention to solve the problems of the prior art described above, to speed up restoration in the event of a recording medium failure, and to smoothly perform normal input/output requests during restoration. The object of the present invention is to provide a method for recovering a recording medium from a failure.

The present invention provides a storage medium failure recovery method for an electronic computer including a plurality of storage media and a central processing unit that writes/reads data to/from the plurality of storage media, in which data to be recorded is distributed to different storage media. a means for recording the
A means for sequentially restoring and reading data recorded on a destroyed recording medium from distributed recording media when a failure occurs, and restoring writing to the destroyed recording medium; The present invention is characterized in that it has a means for performing normal write/read processing on a portion of the recording medium currently in use and on a recording medium other than the recording medium on which restoration reading is currently being performed.

[Example] Hereinafter, a failure recovery method for a recording medium according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a schematic diagram showing an embodiment of the failure recovery method for a recording medium according to the present invention.4 In this embodiment, data 1 to data 10 are stored on five disks 21. 22...25 are recorded in a distributed manner and in duplicate.

The central processing unit 11 is capable of writing/reading data to/from each section of the disks 21, 22, . . . , 25.

In addition, when a failure occurs, the central processing unit 11 sequentially restores and reads data 1 to 4 recorded on the destroyed disk, for example, the disk 21, from the disks on which it was recorded in a distributed manner. It has a function of performing restoration writing on the disk 21 that was previously saved. If data 1 to data 4 recorded on the disk 21 are destroyed, data 1 is restored and read from the disk 22 and transferred to the disk 2.
Data 2 is restored and read from the disk 23, and restored and written to the disk 21. In this way, the central processing unit engineer 1 performs the following steps:
From the same machine, up to data 4, restored and read, disk 2
Perform restoration writing to 1.

The central processing unit 11 further has a function of performing normal write/read processing on a portion of the disk that is being restored and read and on disks other than the disk that is currently being restored and read.

As described above, in the recording medium failure recovery method according to the present invention, for example, while data 1 is being restored and copied, the output request 31 for data 1 on the disks 21 and 22 becomes a write process during restoration.

However, data other than data 1 on the disk 22 is data 5.
Output requests 31 for 6.7 and other disks 23.24.25 are processed normally.

[Effects of the Invention] As explained above, the present invention enables high-speed restoration in the event of a disk failure by distributing file contents onto different media and recording them multiplexed. This has the effect of allowing input/output requests to be executed smoothly.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram showing an embodiment of a failure recovery method for a recording medium according to the present invention. 1 to 10: Data 11: Central processing unit 21 to 25: Disk 31: Output request

Claims (1)

[Scope of Claims] A failure recovery method for a recording medium in an electronic computer including a plurality of recording media and a central processing unit that writes/reads data to/from the plurality of recording media, comprising: storing data to be recorded on different recording media; and a method for sequentially restoring and reading the data recorded on the destroyed recording medium from the distributed recording medium when a failure occurs, and restoring writing to the destroyed recording medium. and means for performing normal write/read processing on a portion of the recording medium on which the restored reading is being performed and on a recording medium other than the recording medium on which the restoring reading is currently being performed. A failure recovery method for recording media that enables high-speed recovery.