JPS6257041A - Recovery system of data - Google Patents

Abstract

PURPOSE:To prevent the data update processing capability from being lowered because of data restoration by providing a time information fetch section, a couple of logic blocks, an update completion table, a file I/O control means and an update end control means. CONSTITUTION:In applying data update to logic blocks 41, 42 just after file initialization by a process 1, a file I/O control means 8 reads the block 3 from the file 2 to compare the blocks 41, 42 with time information in added information 51. As a result, when the time information is coincident and it is decided that no up-to-date data exists, the information 51 is generated via a time information fetch section 6. Then, the update data from the process 1 and the information 51 are set to the block 42, the block 3 is written in the file 2 to complete the data update. Then, the present time information fetched from the fetch section 6 is registered in an update completion table 7 as the completion time information via an update end control means 9. Thus, the data update processing capability is not lowered for data restoration.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a data recovery method in a computer processing system, and more particularly to a data recovery method for restoring data in a file to its past state.

[Conventional technology]

Conventionally, there are two methods for recovering this type of data: one is to save the unupdated data to a separate file and run the data recovery process when data recovery becomes necessary, and the other is to save the updated data to a separate file. There is a method in which the updated data in the separate file is written back to the original file when the data update is completed, and when data recovery is necessary, the separate file is deleted.

In addition, in a personal computer or the like, the entire file is simply saved in advance in a separate file by the user's operation, and when data recovery is required, the entire file is simply written back from the separate file.

[Problem that the invention seeks to solve]

The conventional data recovery method described above has the following problems.

The method of storing the unupdated data in a separate file requires processing to manage the separate file, write out the unupdated data to a separate file, and perform data recovery.In order to prepare the information necessary for data recovery, This reduces data update processing performance, and requires a lot of time and special processes for data recovery.

In addition, in the method of storing updated data in a separate file, it is necessary to manage the separate file and write the updated data to the original file when the data update is completed.
The performance of data update processing is reduced to eliminate the need for data recovery.

Furthermore, with the method in which the entire file is saved in a separate file by the user's operation, the user must save the file and write the saved file back to the original file during data recovery. However, it is inferior to the above-mentioned Niga method in terms of reliability because it imposes a heavy burden on both the user and the computer, and data recovery is not automated.

In view of the above-mentioned points, an object of the present invention is to provide a data recovery method that does not reduce data update processing capacity for data recovery.

[Means for solving problems]

The data recovery method of the present invention is a data recovery method for a computer processing system in which a file that is not updated simultaneously by a plurality of processes and a process that updates this file include a time information importing unit for obtaining time information; It is provided in a block, which is a data input/output unit in the file, and one of the blocks stores the latest updated data and additional information including time information obtained from the time information import unit, and the other stores the data of one generation ago. a pair of logical blocks for storing additional information including update completion data or the latest update data and time information obtained from the time information acquisition unit; and a file for storing completion time information of the latest update completion data when referencing and updating data. A logical block having the latest update completion data by comparing the completion time information in the update completion table with the time information in the additional information when referencing and updating data. and a file I/O control means that makes data recovery unnecessary by referring to and updating the latest updated data even if there is incomplete data updating, and the time information importing unit when the data updating is completed and update completion control means for obtaining completion time information from the update completion table and registering the information in the update completion table.

〔Example〕

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

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

The data recovery method of this embodiment consists of a process l that has a file update function, a file 2 that does not receive updates to the same block from another process when data is being updated from a certain process, and a file 2 that has a file update function. Block 3 is an element, and a pair of logical blocks 41 and 42 (logical block 41 or 42) is an element of block 3.
(the configuration of which is shown in more detail in logical block 5), additional information 51 and data 52 that constitute logical block 5, time information import unit 6, update completion table 7, and file I
It is composed of /O control means 8 and update completion control means 9. Note that in FIG. 1, arrows indicate enlarged displays of constituent elements, and white arrows indicate data flows.

Referring to FIG. 2, the processing in the file I/O control means 8 includes a block reading step 20, a latest data determination step 21, a latest data presence determination step 22, an update completion determination step 23, and an additional information import step 24.
, a data update step 25 and a block write step 26.

Referring to FIG. 3, the process in the update completion control means 9 consists of a completion time information acquisition step 30 and a time information registration step 31.

Next, the operation of the data recovery method of this embodiment will be explained with reference to the drawings.

When process 1 updates data to logical block 5 immediately after file initialization, file I/O control means 8 operates in response to a request from process 1.

The file I/O control means 8 transfers the block 3 containing the data to be updated to the file 2 in the block reading step 20.
In the latest data determination step 21, the time information in the additional information 51 of the logical blocks 41 and 42 in the block 3 is compared.

This comparison result is checked in the latest data presence/absence determination step 22, but since both logical blocks 41 and 42 have not been updated, their time information matches.
It is determined that there is no latest data. Therefore, in the tertiary additional information importing step 24, the current time information is imported from the time information importing section 6 and additional information 51 is created.

Next, in a data update step 25, the updated data from the process 1 and the additional information 51 are set in the logical block 42, for example, and if there is no latest data, the updated data and the additional information 51 from the process 1 are set in the logical block 42.
Update data and additional information 5 for either 1 or 42
I), and in the block write step 26, block 3 is written to file 2.

When the process 1 completes updating the data in this manner, the update completion control means 9 operates. The update completion control means 9 takes in the current time information from the time information take-in section 6 in a completion time information take-in step 30, and registers this time information in the update completion table 7 as completion time information in a time information registration step 31.

When process 1 updates data that has already been updated, file I/O control means 8 operates in response to a request from process 1.

The file I/O control means 8 transfers the block 3 containing the data to be updated to the file 2 in the block reading step 20.
In the latest data determination step 21, the time information in the additional information 51 of the logical blocks 41 and 42 in the block 3 is compared.

Now, assuming that the time information in the additional information 51 in the logical block 41 is newer than the time information in the additional information tll151 in the logical block 42, this logical block 41 is determined to be the logical block having the latest data.

This determination result is checked in the latest data presence determination step 22, and since the latest data is present, control moves to update completion determination step 23.

In the update completion determination step 23, the completion time information in the update completion table 7 is compared with the time information in the additional information 51 in the logical block 41 to determine whether it is newer. If the time information in the additional information 51 in the logical block 41 is newer than the completion time information in the update completion table 7, it is determined that the update of the logical block 41 has been completed.

If the time information in the additional information 51 in the logical block 41 is not newer than the completion time information in the update completion table 7, the update of the logical block 4I is incomplete, and the data update of the logical block 42 is completed. It is determined that the

Subsequently, in an additional information importing step 24, current time information is imported from the time information importing section 6, and additional information 51 is created.

Next, in the data update step 25, the updated data is generated based on the update data from the process l and the data 52 in the logical block 41 or the logical block 42 in which the update completion data was previously determined in the update completion determination step 23. create.

Next, assuming that the logical block 42 is a logical block having the latest updated data, the updated data and additional information 51 are set in the logical block 41, and block 3 is written to the file 2 in a block writing step 26.

Further, assuming that the logical block 41 is a logical block having the latest updated data, the updated data and additional information 51 are set in the logical block 42, and the block 3 is written to the file 2 in a block writing step 26.

In this manner, when the process 1 completes updating the data, the update completion control means 9 operates. The update completion control means 9 takes in the current time information from the time information take-in section 6 in a completion time information take-in step 30, and registers this time information in the update completion table 7 as completion time information in a time information registration step 31.

〔Effect of the invention〕

As explained above, the present invention places the updated data and the updated data in the I/O/O0 block.
It has the following effects.

Even if the data in the file has not been updated yet, there is no need for data recovery processing by a separate process.

Further, since no extra separate file I/O or separate block I/O is required to prepare for data recovery when updating data, the performance of data updating is not impaired.

Furthermore, by adding time information to data and placing time-varying data in different logical blocks, data recovery is unnecessary through simpler and clearer processing compared to conventional data recovery methods.

In addition, since the interface with existing functions of O3 is simple, there is a high degree of freedom as to which layer within O3 it can be placed. For example, it is even possible to place it as firmware in an intelligent floppy disk unit.
It can be applied to various file I/Os.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram showing one embodiment of the present invention, and FIG.
FIG. 3 is a flowchart showing the processing in the file I/O control means shown in FIG. 1. FIG. 3 is a flowchart showing the processing in the update completion control means shown in FIG. In the figure, 1...Process, 2...File, 3...Block, 5...Logic brick, 6...Time information import unit, 7... ...Update completion table, 8...File I/O control means, 9...Update completion control means, 20...Block reading step, 21...Latest data determination step, 22...Latest data presence/absence determination step, 23...Update completion determination step, 24...Additional information import step, 25...Data update step, 26...Block writing step, 30 . . . Completion time information acquisition step, 31 . . . Time information registration step, 41. 42. Logic block, 51 . . . Additional information, 52 . . . Data. ￥k1 Figure 2 Figure 3 Updated hanging/control means 9

Claims (1)

[Scope of Claim] A data recovery method for a computer processing system in which there is a file that is not updated simultaneously by multiple processes and a process that updates the file, comprising: a time information import unit for obtaining time information; Provided in a block that is a data input/output unit, one of which stores the latest update completion data and additional information including time information obtained from the time information acquisition section, and the other one stores update completion data of one generation ago. or a pair of logical blocks for storing the latest updated data and additional information including time information obtained from the time information importing unit; A logical block having the latest update completion data is determined by comparing completion time information in the update completion table and time information in the additional information when referencing and updating data. , a file I/O control means that makes data recovery unnecessary by referencing and updating the latest updated data even when data has not been updated; A data recovery method comprising: update completion control means for obtaining information and registering it in the update completion table.