JPH02287644A - Master file updating system - Google Patents

Abstract

PURPOSE:To execute the restoration at the time of fault from an updating journal in a short time and to shorten the system restoration time by constituting this system so that data which is enough by only a reference uses a master file, and data which requires updating uses an auxiliary master. CONSTITUTION:In the case a master file 2 in which separate information is accumulated is updated in a real time, an auxiliary master 3 is updated, while referring to the file 2 in accordance with updating instruction data 5, and its updating instruction data 5 and a result of updating are accumulated in an updating journal 4. Subsequently, a retrieval module 6 retrieves the contents of the master 3 and the file 2 in accordance with an instruction of retrieval instruction data 9. Also, a master updating module 7 stores the contents of the master 3 updated by the module 7 into the file 2 after the regular work is ended, and the stored data is erased from the master 3 and the journal 4. In such a way, the updating data quantity existing in the master 3 can be decreased, therefore, the restoration at the time of fault can be executed from the journal 4 in a short time, and the restoration of a fault can be executed quickly.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a master file updating method, and particularly to a master file updating method that can quickly restore a master file that has failed due to a power outage or the like in an information processing system. .

[Conventional technology]

Conventionally, in an information processing system, a master file update method for updating a master file in which individual information such as product information is accumulated in real time is used as a decoy as shown in FIG. A system is adopted in which the update information is directly updated and the update information is accumulated in the update journal 4. As shown in FIG. 3, the contents of the update journal 4 are erased after the contents of the master file 22 are periodically saved in another recording medium 23 (for example, after each day's work is finished). Additionally, a failure in the master file 22 generally occurs when the system is stopped due to a power outage while updating the master file 22, but if a failure occurs in the master file 22, As shown in the figure, the master file is restored from the recording medium 23 that stores the contents of the latest master file, and then the update journal 4 is used to store the contents from the time when the master file 22 fails. The method is to update the restored master file.

[Problems that the invention attempts to solve]

However, with the conventional master file update method described above, the larger the amount of data in the master file, the longer it takes to restore the master file, and operations that attempt to reference or update the master file must be stopped during that time. There is a drawback.

[Means to solve the problem]

In the master file update method of the present invention, as shown in FIG. 5, an auxiliary master 3, which is an auxiliary file exclusively for updating, is prepared separately from the master file 2, and all update operations are performed on the auxiliary master 3. A master file failure usually occurs when the system stops during a master file update operation. If the system stops while referring to the master file, this will not cause a problem with the master file. When updating the auxiliary master 3, the master file 2 is only referenced, and all update operations are performed on the auxiliary master 3, thereby making it possible to localize file failures only to the auxiliary master 3. Further, in the event of a failure in the auxiliary master 3, recovery can be achieved only by restoring the auxiliary master 3 from the update journal 4, as shown in FIG. Furthermore, by minimizing the amount of update data for the auxiliary master 3, the recovery time can be shortened. To do this, as shown in FIG. 7, the contents of the auxiliary master 3 are referred to the update journal 4, the master file 2 is updated, and after the update, the contents of the auxiliary master 3 and the update journal 4 are deleted. do. This process is performed periodically (for example, after each day's work is finished).

〔Example〕

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

FIG. 1 is a block diagram of an embodiment of the present invention. In FIG. 1, 1 is an update module, 2 is a master file, 3 is an auxiliary mask, 4 is an update journal, 5 is update instruction data, 6 is a search module, 7 is a mask update module, 8
is a recovery module, 9 is search instruction data, updates the auxiliary master 3 while referring to the master file 2 according to the update instruction data 5, and sends the update instruction data 5 and the update result to the update journal 4, which is used to accumulate update data. Accumulate.

The search module 6 follows the instructions of the search instruction data 9.
Search the contents of the auxiliary master 3 and master file 2. The mask update module 7 normally stores the contents of the auxiliary master 3 updated by the mask update module 7 in the master file 2 after the end of business, and erases the stored data from the auxiliary master 3 and the update journal 4.

Through this processing, the amount of update data existing in the auxiliary master 3 can be reduced, and rapid failure recovery can be performed. The recovery module 8 is activated when the auxiliary master 3 has failed, and updates the auxiliary master 3 according to the update instruction data 5 and update results accumulated in the update journal 4 after the auxiliary master 3 is initialized.

[Detailed Description of the Invention As described above, the present invention uses a master file for data that only needs to be referenced so that normal system failures do not affect the master file, and data that needs to be updated is stored in an auxiliary mask. This has the effect of speeding up system recovery by periodically copying this content to the master file to reduce its capacity, and by allowing recovery from failures to be performed from the update journal in a short time. There is.

[Brief explanation of drawings]

Figure 1 is a block diagram of an embodiment of the present invention, Figure 2 is a diagram for explaining a conventional master file update method, and Figures 3 and 4 are diagrams for explaining conventional master file storage and failure recovery, respectively. 1, 5, 6, and 7 are diagrams for explaining updating of the auxiliary mask, recovery from failure of the auxiliary mask, and transfer from the auxiliary mask to the master file, respectively, in the present invention. It is. 1... Update module, 2... Master file, 3... Auxiliary master, 4...
Update journal, 5...Update instruction data, 6.
...Search module, 7...Master update module, 8...Recovery module, 9...
...Search instruction data.

Claims (1)

[Claims] In an information processing system that updates a master file that accumulates individual information in real time, the master file is used only for reference during normal processing, and a separate update-only file is used for processing that requires updating. An update journal is output, and the master file is updated using the update-only file and the update journal at specific times when normal processing is not performed, and when the update-only file fails, the update journal is used to update the update-only file. A master file update method characterized by file recovery.