JPS6365548A - Backup system for system data - Google Patents

Abstract

PURPOSE:To automate a double storage of a data by referring to a backup table by a sub-job for operating in the course of a holding mode of a main job which has been started, and writing the data block concerned of a user data file to a backup data file. CONSTITUTION:While amain job 1 is processing a user data in an execution mode, a block which necessitates a double storage in a user data is registered in a backup table 3 by a flat, etc. On the other hand, by setting so that a sub- job 2 starts to operate as soon as the main job 1 becomes a holding mode, whenever the execution of the main job 1 becomes a pause for the present, the data block concerned of a user data file 4 is written to a backup data file 5 by the sub-job 2. In this way, a double storage of the user data without troubling an operator's operating consciousness, and also, without considering especially an application program, and without decreasing a processing speed of a machine.

Description

[Detailed Description of the Invention] [Summary] The present invention discloses a technology in which, in a system data bank-up method, data is automatically double-saved by a sub job that operates during the standby mode of the activated main job. It is something to do.

[Industrial application field]

The present invention relates to a system data backup method,
In particular, the present invention relates to a system data backup method that automatically double-saves data.

[Conventional technology]

BACKGROUND ART Double storage (backup) of data has traditionally been extremely important in computer management and operation.

Although this is self-evident considering that the essential function of a computer is to store and process data, it is difficult to say that this is always done smoothly in actual use.

[Problem that the invention seeks to solve]

With office automation computers and terminals that have become increasingly popular in recent years, data storage, especially double storage, is seen as an additional task other than the original work and is not performed, partly because there is no dedicated computer staff. In many cases, a system failure would cause major confusion, and even if the data was saved, it would be done after work hours, resulting in a heavy burden on the workers. On the other hand, in the case of large online machines that do not require human intervention, the application program generally assumes two files, uses one at all times, and if a problem occurs with one, it uses the other. However, this means that it is necessary to write double save and create a substitute section in the program, which causes the problem that program execution is interrupted each time. are doing.

The present invention has been devised in view of these problems, and it is possible to do so without impairing the operator's operational consciousness (and without reducing the processing speed of the machine, without giving special consideration to the application program). The purpose of the present invention is to provide a system data backup method that allows dual storage of user data.

[Means for solving problems]

FIG. 1 is a block diagram showing the principle of the present invention.

In the present invention, the means taken to solve the above problems is a system data backup method in which data management is performed by the main job 1 which has an execution mode and a standby mode. A system data backup method is used in which the sub job 2 operating in standby mode refers to the backup table 3 and writes the corresponding data block of the user data file 4 to the backup data file 5.

Note that, in this specification, a main job is data processing that is directly required by a user, and a sub job is a process that is incidental thereto.

[Effect]

In the present invention, while the main job 1 is processing user data in execution mode, blocks in the user data that require double storage are registered in the backup table 3 using flags, etc. , Set (start) sub job 2 to run immediately when main job 1 enters standby mode.
Then, each time the execution of the main job 1 finishes, the sub job 2 writes the corresponding data block of the user data file 4 to the backup data file 5.

Note that the method for activating the sub-jobs is not particularly limited; a library dedicated to standby jobs may be provided, each job may be identified, or a command may be issued by a higher-level CPU.

〔Example〕

Hereinafter, the present invention will be explained in detail with reference to examples and drawings.

Configuration example of data management according to the present invention FIG. 1 is a configuration diagram schematically showing an example of a system data backup method in which the present invention is implemented together with the principle of the present invention. In FIG. 1, the system data backup method is such that the secondary zip 2 operates during the standby mode of the main job 1, which has an execution mode and a standby mode, and by referring to the backup table 3, the user data file 4 is Back up the corresponding data block.
It is designed to be written to data file 5.

Main job 1 is executed in user application 6 (R
UN) is activated when a command is given, for example, the first
The word processor is started with the command (A), and the second command (B)
If the accounting is set to be started in response to these instructions, the processing means for each main job, that is, the word processor (
A) or the accounting machine (B) will operate and process the daily allowance data block (B-1 to B-4) or (B-7 to B-8) of the user data file 4. on the other hand,
The processing means of the sub job 2 operates based on a program such as the standby job library 7, but regarding its operation, it is determined whether the main job 1 is in the execution mode or the standby mode. A task management table 8 is provided, and by referring to the table 8, all activated main jobs are placed on standby (WAI).
T), the main job or the standby job (W) activated in advance by O8 is operated.

Then, when the activated standby job (W) in operation is a data backup zip, the sub job processing means refers to the backup table 3.

The backup table 3 is a table that manages pages of each data block (B-1 to B-8) of the user data file 4, and is flagged as "0" by default, and then the main zip 1 When reading or writing data, the required data blocks (e.g. B-3 and B-4)
is rewritten, the flag of the corresponding data block (B, 3. B-4) of backup table 3 is also set to "1".
This flag "1" has the meaning of a bank up request, and when the standby job (W) is a backup job and table 3 is referenced, the corresponding data block (B-3 , B-4), and in sub job 2, user data file 4
The data block (B-3, 8-4) of 8 is written to the backup data file 5, and the flag of the data block is rewritten to "0". If all data blocks are not rewritten during the standby mode, the flags of the remaining data blocks are set to 1'.
'' and will be processed next time. Therefore, whether or not the required data has been copied is constantly checked in the backup table 3, and double saving is automatically processed without the operator's knowledge. and will not degrade system performance.

In the above embodiment, it is assumed that the waiting job library 7 is given an extremely low priority and the sub job 2 is operated in WAIT by referring to the task management table 8. However, in general, each library 6.7 and the task management table 8, as shown by the dotted line in FIG.
WAIT of main job 1 directly by multiple upper CPUs 9
The sub job 2 may be operated by a command from the CPU 9 each time.

Flowchart of the method of the present invention FIG. 2 is a flowchart of the above configuration example.

In Figure 2, the flow starts when the main job is started, and when the user data file is rewritten during data management, the flag of the item corresponding to the corresponding data block in the backup table is changed to "
1”.

Eventually, when the main job enters standby mode, the activated sub job is operated, searches the backup table, and when it finds the data block with flag 1'', writes the data of that block to the backup data file, Rewrite the backup table flag to "O".

When the main job is no longer on standby (WAIT) or when all blocks have been searched and rewritten, the flow of the sub job stops the loop and returns to the flow of the main job. Note that this flowchart only shows the data management-related portion of the present invention, and is not an interface between the main job and the sub-job. Further, details of the main job are omitted.

Example of configuration of data medium of the present invention Floppy disks are commonly used as the medium for the above-mentioned backup data files, but in the configuration of low-cost equipment such as office computers and personal computers, magnetic disks can be used in addition to floppy disks. Tapes, especially cassette tapes, are often used. FIG. 2 is an explanatory diagram showing a backup method when a medium that cannot be randomly accessed, such as a magnetic tape, is used in the present invention. In FIG. 2, the magnetic tape ■, which is a strip-shaped medium, is
It has a length sufficient to accommodate the capacity of the data file 4, and is divided into usage sections for each data block in the tape traveling direction. A header section 11 containing a block number and write time is provided at the beginning of each usage section in the forward direction using the entire width or part of the tape, a backup data section 12 is provided at the center, and an end section is provided at the end. 13 are arranged. If data with the same block number is recorded on the same tape, their writing times are compared and the one closest to the current one is considered valid. In order to be able to perform further writing even after the magnetic tape "L"- has been fully recorded, the recorder is configured to be reversible, and the usage section where the same block number is recorded while running in the opposite direction. Either search for and overwrite that part, or configure the magnetic tape endlessly and repeatedly overwrite the used sections in which the same block number is recorded. By using such a magnetic tape, in the present invention, bank up processing similar to that of a random access device can be performed using a general magnetic tape.

〔Effect of the invention〕

As described above, according to the present invention, the application
To provide a system data backup method that allows double saving of user data without reducing machine processing speed without special considerations in the program, and of course fundamentally prevents forgetting to save data. be able to.

[Brief explanation of drawings]

FIG. 1 is a schematic block diagram of a data management method according to the present invention, FIG. 2 is a flowchart of the above method, and FIG. 3 is an explanatory diagram of an example of a data medium used in the present invention. 1; Main job, 2; Sub-job, 3; Backup table, 4: User data file, 5; Backup data file, 6; Application library, 7S standby job library, 8; Task management table, 9; CPU. Y: Magnetic tape. FIG. 1 FIG. 2 FI-style nine configuration ports of data management system according to the present invention

Claims (1)

[Claims] A method for backing up system data processed by a main job (1) having an execution mode and a standby mode, the system data being backed up by a sub job (2) that operates during the standby mode of the started main job (1). ) by referring to the backup table (3), the corresponding data block of the user data file (4) is transferred to the backup data file (5).
) system data backup method.