KR100800044B1 - Method for automatic checkpoint file management - Google Patents

Abstract

A method for automatically managing checkpoint files is provided to keep only the optimal checkpoint files in a restricted storage space by selecting and removing only the terminated tasks after checking a current executing state of the task in the used checkpoint file. It is checked whether the current checkpoint file uses the storage space over a predetermined setting value(S10). The current checkpoint file is searched if the storage space is over the setting value(S20). A task ID of the searched checkpoint file is extracted(S30). The checkpoint file of the terminated task is removed by determining whether the task corresponding to each extracted task ID is currently executed(S60). A checkpoint file management operation is terminated if the storage space of the checkpoint file is below the setting value.

Description

Method for automatic checkpoint file management}

1 is an exemplary diagram illustrating the use of a general checkpoint file as an example;

2 is a flowchart illustrating a checkpoint file automatic management method according to the present invention.

The present invention relates to a method for automatically managing a checkpoint file and a computer-readable recording medium having recorded thereon a program for realizing the same. More particularly, the present invention relates to whether a checkpoint file is executed or not. Checkpoint file automatic management method that manages to maintain only the optimal checkpoint file in the limited storage space by checking and determining unnecessary work terminated based on this, and the computer for recording the program for realizing it A readable recording medium.

When designing a computer system, reliable technologies in computer systems are widely used to prepare countermeasures in advance in case of an error, so that normal operation can be continued without an effect even when an error occurs.

An example of such an in-system trust technique is a checkpoint file. A checkpoint file checks the memory state and writes it to a file while the computer program is running. It is a technology (or storage device) that allows a user to start again from a stored part. More specifically, if the work time on a basic file is several hours to several days in a row, a certain period of work is performed in order to prevent a situation in which all the work is returned to the origin due to the destruction of the file due to an accident or failure. By saving the intermediate result of the work in progress as a file, that is, a checkpoint file, it is possible to restore the work file in progress based on the last saved checkpoint file when a failure occurs.

FIG. 1 shows an example of a method for using a general checkpoint file. As shown in FIG. 1, the task (a) has a period of checkpoints A, B, and C until the end of execution. Save your work (a, b, c). If a task is interrupted due to a failure during the execution of the task (a), the task is restored based on the last saved checkpoint file. For example, the task result is saved from the point A to the point in time. If the job is interrupted due to occurrence, then the job will recover the job (a) based on the checkpoint file (a) stored at the time A.

However, the conventional checkpoint file is configured to be deleted together automatically when the task ends. This means that if the task is abnormally terminated due to a failure, the task cannot be recovered because the checkpoint file for the task is deleted at the same time as the task is terminated.

Therefore, there should be a search for an improved reliability technique of a computer system that improves the problem of the conventional checkpoint file which is implemented to be automatically deleted at the end of the operation.

Accordingly, the present invention has been made to solve the above-mentioned conventional problems, the present invention is to check the current execution status of the operation in the checkpoint file in use to select only the terminated task and delete the optimal Its purpose is to provide an automatic checkpoint file management method that maintains only checkpoint files.

Checkpoint file automatic management method according to the present invention for achieving the above object comprises a check step of determining whether the currently used checkpoint file is using a storage space of a specific set value or more; A search step of searching for a checkpoint file currently in use when the checkpoint file usage is equal to or greater than the specific setting value as a result of the determination; ID extraction step of extracting a job ID for the searched checkpoint file; And a deletion step of determining whether a job corresponding to each of the extracted job IDs is currently being executed, and performing a checkpoint file deletion for the job whose execution has ended.

The present invention also provides a method for automatically managing a checkpoint file, comprising: a checking step of determining whether a checkpoint file currently being used uses a storage space equal to or greater than a specific setting value; A search step of searching for a checkpoint file currently in use when the checkpoint file usage is equal to or greater than the specific setting value as a result of the determination; ID extraction step of extracting a job ID for the searched checkpoint file; And a deletion step of determining whether a job corresponding to each of the extracted job IDs is currently being executed, and performing a checkpoint file deletion for the job whose execution has ended; a computer-readable recording medium having recorded thereon a program for realizing It is starting.

Hereinafter, a method for automatically managing a checkpoint file according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a checkpoint file automatic management method according to the present invention.

Automatic management of the checkpoint file can be deployed only when the usage of the checkpoint file is more than a predetermined value. Therefore, the automatic checkpoint file management method according to the present invention starts with a process of checking whether the checkpoint file usage is equal to or greater than a predetermined specific value (S10). During the work on the computer system, a checkpoint of a certain time is placed on the work, and the running work is stored in a file on a storage device such as a disk. This is called a checkpoint file. . The storage space for storing such checkpoint files is limited, and if a large number of checkpoint files remain unmanaged, it causes a problem of wasting storage space due to storing the checkpoint file instead of the current execution state. . In addition, you will not be able to create other new checkpoint files. To this end, the automatic checkpoint file management method according to the present invention provides a process for managing a checkpoint file when the checkpoint file usage occupies a predetermined value or more in a corresponding area in a storage area allocated for storing the checkpoint file. Will develop. In this case, the specific value for the checkpoint file usage, which is a key factor of the process development, is a value that is set at the time of program development for realizing the checkpoint file management method.

If the checkpoint file usage does not reach a predetermined value, the checkpoint file stored in the current storage device is determined to be unnecessary and stops the operation for checkpoint file management. However, when it is determined that the checkpoint file usage is equal to or greater than a predetermined value, the checkpoint file currently stored in the storage device is determined as a management target, and the management procedure according to the present invention is performed.

The management procedure for the checkpoint file is as follows.

First, a search for a checkpoint file being used is performed (S20). That is, the checkpoint file stored in the current storage device is searched. For example, a user may input a specific command for searching a checkpoint file under a checkpoint file management environment, and as a result, may be provided with a list of currently used checkpoint files.

When a checkpoint file being used as a result of performing the twentieth step S20 is searched for, a job ID for the corresponding checkpoint file is extracted (S30).

When the task ID extraction process is completed, the number of extracted task IDs is checked to determine whether to proceed with a management procedure according to the verification result (S40). In detail, the number of extracted job IDs is checked, and if the extracted job ID does not exist, the operation for checkpoint file management is stopped. If the number of extracted job IDs exists, the following operation is performed for checkpoint file management.

If it is determined that the extracted job ID exists, as the next operation, it is determined whether the corresponding job for the extracted job ID is currently running (S50), and the checkpoint file allocated to the job is deleted based on the determination result. (S60). More specifically, in the extracted job ID, check whether the corresponding job for the job ID is currently executed, and if the corresponding job for the job ID is currently running, skip the procedure for deleting the checkpoint file for the job. Proceed with checking whether the task is executed for the task ID. If the corresponding job for the job ID is currently terminated and is not used, the checkpoint file used for the job is deleted. At this time, the step of deleting the checkpoint file or omitting the deletion procedure is to repeat the process by the number of extracted job IDs.

Each step for managing the checkpoint file may be developed and executed as a program, and may be inserted into a file in a specific driving environment for automatic execution of the corresponding program. For example, the program can be inserted into a crontab file so that it can be executed automatically. In this case, crontab executes a command automatically, and executes the command at the date and time specified by the user. That is, the checkpoint file management method as described above is developed as a program and inserted into the crontab file to be executed so that file management for the checkpoint currently being used can be periodically performed.

The automatic checkpoint file management method according to the present invention as described above is implemented as a program, a computer-readable recording medium, that is, random access memory (RAM), read only memory (ROM), hard disk, optical disk And the like.

The automatic checkpoint file management method has been described with reference to the preferred embodiment. It will be appreciated that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention by those skilled in the art. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation, and the scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent scope are included in the present invention. Should be interpreted as.

As described above, according to the automatic checkpoint file management method according to the present invention, by checking the current execution state of the existing task in the checkpoint file by selecting only the task for unnecessary tasks to delete the checkpoint file, It is possible to maintain only the optimal checkpoint file in the limited storage space and further promote efficient use of the storage space.

In addition, by maintaining and managing the checkpoint file regardless of whether the job is terminated, it is possible to solve the problem of job recovery due to automatic deletion of the checkpoint file at the end of an abnormal job.

Claims (9)

As a way to automatically manage checkpoint files, A checking step of determining whether a checkpoint file currently being used uses a storage space equal to or greater than a specific setting value; A search step of searching for a checkpoint file currently in use when the checkpoint file usage is equal to or greater than the specific setting value as a result of the determination; ID extraction step of extracting a job ID for the searched checkpoint file; And A deletion step of determining whether a job corresponding to each of the extracted job IDs is currently being executed and deleting a checkpoint file for the job whose execution has ended; Checkpoint file automatic management method comprising a. The method of claim 1, The checking step further comprises the step of ending the checkpoint file management operation, if the checkpoint file usage does not reach a predetermined value as a result of the execution. The method according to claim 1 or 2, The ID extraction step further comprises the step of terminating the checkpoint automatic management operation if the extracted job ID does not exist, the task ID extraction method further comprises. The method of claim 1, And the deleting step is repeated as many times as the number of job IDs extracted in the ID extracting step. The method of claim 1, The automatic checkpoint file management method is embedded in an automatic execution file and automatically executed at a designated time. In performing checkpoint file management, A checking step of determining whether a checkpoint file currently being used uses a storage space equal to or greater than a specific setting value; A search step of searching for a checkpoint file currently in use when the checkpoint file usage is equal to or greater than the specific setting value as a result of the determination; ID extraction step of extracting a job ID for the searched checkpoint file; And A deletion step of determining whether a job corresponding to each of the extracted job IDs is currently being executed and deleting a checkpoint file for the job whose execution has ended; A computer-readable recording medium having recorded thereon a program for realizing the same. The method of claim 6, And the checking step further comprises the step of terminating the checkpoint file management operation when the checkpoint file usage does not reach a predetermined value as a result of the execution. The method according to claim 6 or 7, The step of extracting the ID further comprises the step of terminating the checkpoint automatic management operation when the extracted job ID does not exist, as a result of the job ID extraction. The method of claim 6, And the deleting step is repeated as many times as the number of job IDs extracted in the ID extracting step.

Images