JPH07244600A - Method and device for controlling backup file - Google Patents

Abstract

PURPOSE:To make it possible to recover data eve when the data are not backed up by another medium by preparing a status file control file for controlling a stale area and recovering fills from the stale area. CONSTITUTION:A system (a) controls a stale file control part 10 and an operator (g) outputs a recovery request to the control part 10. When a recall controlling controller (b) receives a recall request (4) from the system (a), a stale file registering mechanism (c) refers to (6) an existing MO file control table (d) to prepare (7) a stale file control table (f) based upon the referred information and transfers (9) table information (8) obtained by referring to the table (d) to the controller (b), which executes (10) the recall. Thereby backup data can be generated without requiring copy requesting time for backup and a backup- only medium.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a backup file management method and apparatus for effectively using existing data on an optical disk.

In the present market, a system for collectively managing hard disks (fixed magnetic disks) and optical disks as a file collective management system, in which files that are frequently processed are stored in the hard disk and files that are less frequently processed are stored in the optical disk. Is coming out. This takes advantage of the characteristics that the hard disk has a high processing speed, the optical disk is slow, but the storage capacity is large, and the control unit of the file batch management system (hereinafter referred to as "system") sets it for each file system on the hard disk side. Check the allowable usage amount, and when it reaches that, refer to the last access date and time of the file on the hard disk side, write from the oldest file to the optical disk side, and delete the file on the hard disk side. ")) Method to enable file saving.

When there is an access request for a file moved to the optical disc side, the system refers to the file management table for moving the optical disc to the system, searches the address on the optical disc from the information, and reads the corresponding file. The process of writing to the hard disk and using it for access execution (hereinafter, this operation is called "recall") is performed. After that, the hard disk is accessed as long as the corresponding file exists on the hard disk.

In the recall process, the file on the optical disk medium after reading is deleted from the management table of the file on the optical disk at the same time as writing on the hard disk, but the actual data is not deleted and is set as an invalid area. After that, it is left until it is erased all over the medium at any time (referred to as "stale area").

The present invention relates to a file management system having such a function.

[0006]

2. Description of the Related Art Conventionally, backup has been used as a means for improving the reliability of data. This is essential in order to guarantee the safety of data even if the accuracy of the external storage device is improved.

Optical disks and magnetic tapes are used as backup storage media. As this backup method, currently, the operator himself writes down the backup date, file name, etc. for backup management, or the time management table (Coolong tab) that the system automatically backs up at the time preset by the operator. There was something to do.

When a file on the hard disk is lost due to a system down or database loss, the data that has been backed up in advance is copied to another medium for recovery (recovery). )I do.

Further, when it is desired to recover a file at a more detailed time, the system performing the transaction processing uses the transaction file containing the data for updating the master file. I was doing a recovery.

[0010]

The above-mentioned conventional ones have the following problems. For backup, the operator prepares for recovery by performing backup work on a separate medium dedicated to backup. However, if the operator forgets to take a backup, there is no way to restore it and there is no choice but to give up. Also, regarding the transaction processing, if the master file is lost, it is impossible to recover the file.

Further, even when the system automatically backs up at a predetermined time, there may be a case where there is no backup file due to operator's setting error or forgetting to set.

In any case, it is necessary to carry out the backup process manually, requiring a time for copying for backup, and requiring a backup location on the medium.

It is an object of the present invention to solve the above-mentioned conventional problems and to make it possible to recover data by using the stale area even if the operator himself / herself has not backed up to another medium when the hard disk is lost. To do.

[0014]

The present invention has the following structure in order to solve the above conventional problems. FIG. 1 is a diagram for explaining the principle of the present invention. The system a controls the stale file management unit 10, and the operator g
A recovery request is issued to the stale file management unit 10.

The stale file management unit 10 includes a recall control controller b, a stale file registration mechanism c,
An optical disk (MO) file management table d, a migration control controller e, a stale file management table f, and a recovery control controller h are provided.

The recall control controller b is a controller that receives a recall request and executes a recall. The stale file registration mechanism c creates a stale file management table f for a file for which a recall request has been made, searches the stale file management table f for information on a file for which a recovery request has been made, and transfers this to the recovery control controller. MO file management table d
Is a table for managing files on the optical disc.
The migration control controller e is a controller that receives the migration request, creates the MO file management table d, and executes the migration.
The stale file management table f is a table for managing the stale area of the optical disc. The recovery control controller h is a controller that receives a recovery request and executes recovery.

[0017]

The operation of the present invention will be described with reference to FIG. In the figure, (1) to (16) show each processing.

When the migration control controller e receives the migration request (1) from the system a, the migration control controller e executes the migration (3) and creates the MO file management table d of this migration file (2). .

When the recall control controller b receives the recall request (4) from the system a, the recall control controller b passes the registration request to the stale file management table f to the stale file registration mechanism c (5). The stale file registration mechanism c refers to the existing MO file management table d (6), creates the stale file management table f based on the information (7), and
The table information (8) referring to the MO file management table d is transferred to the recall controller b (9). Then, the recall control controller b executes the recall (10).

When the recovery control controller h receives the recovery request (11) from the operator g, the recovery control controller h passes a reference request to the stale file management table f registered in the stale file registration mechanism c ( 12). The stale file registration mechanism c searches the file for which the recovery is requested from the information in the stale file management table f (13) (1
4) The information is transferred to the recovery controller h (15). As a result, the recovery control controller h
Performs recovery (16).

As described above, the stale file management table f for managing the stale area on the optical disk for each generation
By providing a backup file and switching the stale file to a backup file, backup data can be generated without requiring a time for copying for backup and a backup dedicated medium. Therefore, when the data on the hard disk is lost, the data can be restored even if the operator does not back up to another medium.

[0022]

Embodiments of the present invention will be described below with reference to FIGS. In the figure, the same parts as those in FIG. 1 are designated by the same reference numerals.

(1) Description of the First Embodiment FIG. 2 is an explanatory diagram of the first embodiment. In FIG. 2, M
The O file management table d is a registration example when the files A, B, and C are migrated. For example, the file A of this management table d is 92/12 /
It indicates that 10 bytes are written in the address "1" of the optical disc at the date and time of 20. Further, the stale file management table f is an example when the files A, B, and C are recalled. For example, this management table f
File A is recalled at the date and time of 93/01/01 and migrated to the 10-byte optical disc at the address "1" at 92/12/20.

The file recalled and registered in the stale file management table is to be deleted from the MO file management table. (Description of recall) From system a, for example, file A
(4), the recall control controller b goes through the stale file registration mechanism c (5) and refers to the MO file management table d (6). The stale file registration mechanism c returns the information about the file A in the MO file management table d to the recall control controller b (8) and (9), and creates the information added with the recall date and time to manage the stale file. Register in table f (7). The recall controller b executes the recall (10).
And the files B and C are 93/01 respectively
/ 05, 93/01/10 were recalled.

(Description of Recovery) After that, when a system down or database loss occurs and an operator g requests to recover the file A (11), the recovery control controller h causes the stale file registration mechanism c to execute the stalling. File A in file management table f
The reference request is passed (12). The stale file registration mechanism c uses the file A in the stale file management table f.
(13) and (14), and the information is transferred to the recovery controller h (15). As a result, the recovery control controller h executes recovery from the position information (16).

Further, if there is a request to restore at the time of 93/01/05, the recall date and time in the stale file management table f is referred to, and the information of the file older than 93/01/05, this example In case of, files A and B
Is transferred to the recovery control controller to execute recovery.

(2) Description of the Second Embodiment FIG. 3 is an explanatory view of the second embodiment. As shown in FIG. 3, the update on the hard disk HDD and the hard disk HD are performed.
This is an example of the case where migration and recall between D and the optical disk MO are performed. In this example, file A
0, A1, and A2 are files with the same name, file A1 is file A0, and file A2 is file A.
1 shows the updated version of file 1.

The hard disk HDD normally stores the data for the update of the file as a transaction area, for example, the difference (A1-A0) between the file A1 and the file A0, and is used for updating the master file. In this example, the transaction data is 93
/ 01/02 A1-A0 and 93/01/04 A2-
A1 is stored.

Hereinafter, description will be made according to the time flow of FIG. The file A0 on the hard disk is 92/12 /
20 is migrated to the optical disc MO, and then this file A0 is recalled to 93/01/01, the file A0 is updated to the file A1 on 93/01/02, and the file A1 is renamed to the file A2 on 93/01/04. This file A2 was updated to 93/01 //
Was migrated to the optical disk MO on 05, 93 /
This is the case when the file A2 is recalled on 01/20, and the updated data is stored in the transaction area at the time of the update.

FIG. 4 shows the flow of processing when the processing is performed according to the circumstances of FIG. In FIG. 4, (4) to (2
0) indicates each processing. In FIG. 4, the MO file management table d is provided with information such as the file name, the writing date and time of the file, the address, and the number of writing bytes. In this example, files A0, B0, C0, A2, D0, B3
Is registered. The stale file management table f is provided with information such as recall date / time, file name, writing date / time, address, and number of bytes. In this example, it is registered that the files A0, B0, C0, and A2 have been recalled at their respective recall dates and times.

(Description of Recall) The recalled file is processed in the same manner as in the first embodiment. For example, when the system a receives a recall request for the file A0 (4), the recall control controller b refers to the MO file management table d (5) via the stale file registration mechanism c (5). Steer file registration mechanism c
Returns the information about the file A0 in the MO file management table d to the recall controller b (8) (9), and the recall date / time 93/01/0 is included in the information.
A file with 1 added is created (7) and registered in the stale file management table f. Then, the recall controller b executes the recall (10).

In this example, the files B0, C0, and A2 are also recalled at the respective recall dates and times. The files A0 and A2 are two stale files with the same name as the file name A.
Although 0 and A2 are left in the optical disc and the stale file management table, since the recall date and time are entered in the stale file management table, it is possible to know the old and new of each other and restore at each time point.

(Description of Recovery) For example, the operator g
From 93/01/03 (file A1 in FIG. 3), when the recovery control controller h receives a recovery request to restore the file (11), the recovery control controller h writes the file A to the stale file registration mechanism c.
From the information in the above, the information immediately before the desired date and time (File A
The reference request of 0) is passed (12). The stale file registration mechanism c refers to the information of the file A0 in the stale file management table f (13) (14) and transfers the information to the recovery controller h (15). The recovery control controller h executes recovery (1
6).

Then, the recovery control controller h is
At the same time when the recovery end notification is received, a transaction activation request is issued to the transaction control controller i (17), and the transaction control controller i selects the file A0 from the transaction management table j.
Received the update data information from the file to the file A1 (1
8) (19) Transfer the update data information to the recovery controller h (20). The recovery control controller h executes transaction recovery.

In this way, the transaction data can be used to restore the file at the desired time. (3) Description of Backup File Device FIGS. 5 and 6 are device configuration diagrams of an embodiment of the present invention.

Hereinafter, description will be given with reference to FIG. The file management apparatus of this embodiment is provided with a host 1 such as a workstation equipped with a hard disk HDD and an optical library jukebox 2 which is a library apparatus equipped with an optical disk MO. A control unit 11 for controlling is provided. The control unit 11 is provided with a stale file management unit 10 such as a migration control controller e, a recall control controller b, a stale file registration mechanism c, an MO file management table d, a stale file management table f, and a recovery control controller h. The optical library jukebox 2 includes an optical disk MO and an optical disk MO.
The control unit 21 is provided.

Hard disk H provided on the host 1 side
The DD and the optical disk MO provided on the optical library jukebox 2 side as a subsystem are collectively managed so that migration, recall, and recovery can be performed via their respective control units. Then, the actual movement of signals and data between the hard disk HDD side and the optical disk MO side is performed by, for example, scanning (SCI: Sm).
all Computer System Inter
It is performed via a bus (BUS) using a face.

FIG. 6 shows an embodiment in which the control unit 21 on the subsystem side is provided with a stale file registration mechanism c and a stale file management table f. Hereinafter, description will be given with reference to FIG.

The host 1 is provided with a control unit 11 for controlling the hard disk HDD and the like. The control unit 11 is provided with a migration control controller e, a recall control controller b, an MO file management table d, and a recovery control controller h.

On the other hand, the optical library jukebox 2 is provided with an optical disc MO and a control unit 21 for the optical disc MO. The control unit is provided with a stale file registration mechanism c and a stale file management table f.

Hard disk H provided on the host 1 side
The DD and the optical disk MO provided in the optical library jukebox 2 on the subsystem side are collectively managed so that migration, recall, a search command for a recovery file, recovery, etc. can be performed via their respective control units. It is a thing. As described above, since the stale file registration mechanism c and the stale file management table f are provided on the subsystem side, the load on the host side can be reduced accordingly.

As in the case of FIG. 5, the actual movement of signals and data between the hard disk HDD side and the optical disk MO side is carried out by a bus or the like using scati (SCSI).

Although the hard disk HDD (magnetic disk) is used as the storage medium of the host 1 in the above embodiment, an optical disk can be used. The optical disk MO on the subsystem side can be a random access, rewritable optical disk or magneto-optical disk.

[0044]

As described above, the present invention has the following effects. By providing the stale file management table and switching the stale file to the backup file, backup data can be generated and data can be restored without requiring a backup medium separately from the time required for copying for backup.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is an explanatory diagram of the first embodiment.

FIG. 3 is an explanatory diagram of a second embodiment.

FIG. 4 is an explanatory diagram of a processing flow of the second embodiment.

FIG. 5 is a device configuration diagram in an embodiment of the present invention.

FIG. 6 is a device configuration diagram in which a stale file management table is provided on the subsystem side in the embodiment of the present invention.

[Explanation of symbols]

 a system b recall control controller c stale file registration mechanism d MO file management table e migration control controller f stale file management table g operator h recovery control controller

Claims (4)

[Claims] 1. In a system having a disk for storing frequently processed files, a library device for storing infrequently processed files, and a function for leaving data as a stale area in a medium of the library device when a file is recalled. A backup file management method comprising providing a stale file management table (f) for managing the stale area and recovering a file from the stale area. 2. The backup file management method according to claim 1, wherein transaction recovery is performed using transaction data after completion of recovery of the file. 3. A disk for storing a file with a high processing frequency, a library device for storing a file with a low processing frequency, and a library device having a function of leaving data as a stale area in a medium of the library device during file recall processing. In the backup file apparatus, the stale file management table (f) for managing the stale area is provided on the disk side for storing the frequently-processed file, and the file is recovered from the stale area. 4. The backup file device according to claim 3, wherein the stale file management table (f) is provided in a library device.