JPH04312131A - Job executing control system - Google Patents

Abstract

PURPOSE:To automatically and efficiently perform the data processing shared by plural jobs. CONSTITUTION:The processing data are divided through a computer 1 for automatic production of jobs. A job result control table 8 controls the executing result of each job, and these job executing results are registered by a job result deciding part 7. A job start deciding part 3 starts the next processing when the registration of the job executing result is normally complete based on the register results of the table 8. Meanwhile the following processing operations are discontinued if an error occurs. When a job is started, the quantity of data treated in the next processing is adjusted according to the past processing error occurrence states registered in the table 8.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a job execution management technique, and more particularly to a technique that is effective when applied to job execution management in which a certain amount of continuous data is shared and sequentially processed in a computer system.

0002

2. Description of the Related Art Data processing by computers is performed in various fields, and in general-purpose electronic computer systems, the amount of data processed is often enormous. Particularly, when data must be processed sequentially, the number of man-hours and time involved in processing becomes large.

Moreover, it is generally difficult to process all data at once in one job for the following reasons. That is, 1. In computer centers used by many users,
There is usually a limit to available resources such as CPU time per job, and there is a limit to the amount of processing for one job.

2. There may be errors in data, or operational errors may occur in programs that have just started operation. In such a case, if the amount of processing per job is large, it will take time to analyze the error.

[0005] In order to deal with the above problems, generally,
The following method must be used.

1. Divide data processing into multiple jobs. Then, the next job can be processed using the processing results of the previous job.

2. At the beginning of processing, process a small amount of data, estimate the resources required for processing per unit of data, and after determining that no errors will occur, increase the amount of data processed per job as much as the resources allow. go.

However, when the number of jobs increases due to the division of data processing, it takes more effort to input the jobs. Also,
If workers input each job one by one and check the results, their work efficiency will not improve.

Conventionally, it has been known to control job execution by associating jobs, as described in, for example, Japanese Patent Laid-Open No. 1-147631. According to this method, once several jobs are input, the jobs are processed sequentially in the order in which they are associated. If an error occurs in the processing of a certain job, subsequent processing is canceled, so after correcting the processing error, the remaining jobs can be input and executed in the same way.

0010

SUMMARY OF THE INVENTION However, the above-mentioned prior art has the following problems.

1) The work of dividing the entire process into several jobs and the work of allocating processing data to each job and inputting it into a computer are all done manually.

2) If there is a data error or program operation error, the above 1) must be repeated for data processing after the error is corrected.

[0013] These require a large number of man-hours when the total amount of data is large.

An object of the present invention is to provide a job execution management technique that can solve the problems of the prior art and automatically and efficiently perform data processing by dividing multiple jobs. .

[0015]

[Means for Solving the Problems] That is, the job execution management method according to the present invention processes the data in a computer system that divides a desired amount of continuous data and processes the data by dividing it into several jobs. automatically starts a job to perform the job, determines the success or failure of data processing in the job based on the execution results of each job, records the execution result and success or failure of the job as a data processing result, and records the data processing result of the preceding job. If the success or failure of the preceding job indicates that data processing is possible, the amount of data to be processed by the subsequent job is increased or decreased based on the data processing result. This is what I did.

[0016]

[Operation] According to the job execution management method of the present invention described above, when data processing is executed by a certain job, it is determined whether the job has ended normally or an error has occurred and is recorded as the data processing result. Therefore, before moving on to the next process, refer to past data processing results. For example, if there were no errors in the previous job, increase the amount of data to be processed and start the next job, or vice versa. ,
If an error occurs in a certain job, it is possible to perform control operations such as simply recording the result and not starting the next job.

[0017] Thereby, data processing can be automatically and efficiently performed by sharing multiple jobs.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A job execution management system which is an embodiment of the present invention will be specifically described below with reference to the drawings.

FIG. 1 is a diagram showing an example of the configuration of a computer system in which a job execution management method according to an embodiment of the present invention is executed. Note that the wide arrows in the figure indicate the flow of jobs, and the single-line arrows indicate the flow of control.

In FIG. 1, 1 is a computer. 2 is a processing parameter associated with data processing. The processing parameter 2 specifies, for example, data to be processed.

Reference numeral 3 denotes a job starting control unit, which determines whether or not a job can be processed and processing data for each job. Then, based on the processing parameters 2, a job control language (hereinafter abbreviated as JCL) for the job to be executed is created.

Reference numeral 4 denotes a job input queue, which is a job queue created by the job startup control section 3 and waiting for execution.

Reference numeral 5 denotes a job execution unit of the computer 1, which sequentially takes out jobs from the job input queue 4 and executes them.

6 is data to be processed. In this embodiment, all target processing data is prepared on the computer.

Reference numeral 7 denotes a job result determination section, which checks whether the processing by the job execution section 5 has been completed normally and determines whether the processing was successful or not.

Reference numeral 8 denotes a job result management table in which the success or failure of each job is registered by the job result determination section 7. This table is placed on disk.

Reference numeral 9 denotes a job output queue, in which individual jobs that have been processed form a queue waiting for output.

10 is an output list showing the output results of the job.

FIG. 2 shows the job result management table 8 mentioned above.
An example of the configuration is shown. Job result management table 8
consists of processing data 6 corresponding to each job (in this example, a number assigned to the processing data) and a flag indicating success or failure of the job. This flag indicates that '1' indicates success;
'0' indicates failure. For example, in the example of FIG. 2, job A was unsuccessful, but jobs B, C, D,
E shows that the expected results were obtained.

FIG. 3 is a flowchart showing an example of the operation of the job execution management method of this embodiment.

Hereinafter, according to the flowchart of FIG. 3,
An example of the operation of the job execution management method of this embodiment will be explained.

First, processing is started by inputting processing parameters 2 (step 101), and job result management table 8 is prepared on the disk (step 102).

Next, the job starting control unit 3 determines the range of data to be processed in the job to be executed from now on (step 103). Based on this, JCL was created,
The job is input into the job input queue 4 (step 104).

From the job input queue 4, the job execution unit 5
The job is retrieved and executed (step 105).

After the job is completed, the job result determination unit 7 determines the result (step 106). If the job has ended normally, a flag indicating success is set in the job result management table 8. Otherwise, a flag indicating failure is set (step 107).

Note that the output results of each job are output to the job output queue 9.

Next, the job startup control unit 3 refers to the job result management table 8, confirms the result of the job, and determines whether to perform subsequent data processing (step 1).
08). If the preceding job is unsuccessful, no subsequent processing is performed. Here, the job result management table 8 is left on the disk. In the case of success, if there is unprocessed data in the processed data 6, data processing is continued (step 109), and the process returns to determining the next target processing data (
Step 103).

The above process is repeated, and when there is no more data to process, the process ends (step 110).

FIG. 4 is a flowchart showing an example of a processing procedure for determining the amount of data to be processed in the job startup control section 3. An example of the processing procedure will be described below according to the figure.

First, referring to the job result management table 8, the CPU time required for processing per unit data is calculated from past processing results (steps 201 and 202).

Next, search for success or failure of past data processing,
Check whether an error has occurred (step 203)
). The number of past processes to be searched is specified by the process parameter 2. If an error occurs, the processed data is not increased and remains the same as the previous processing amount.

If no error has occurred, then the amount of data increase is tentatively determined (step 204). The rate of increase is specified by processing parameter 2. Here, using the CPU time per unit data described above, check whether the tentatively determined data amount can be processed within the CPU time limit (
Step 205). If processing is not possible, adjust the amount of data (step 206) and end the processing (step 20).
7).

As explained above, according to this embodiment,
After inputting the processing parameters 2, there is no need for manual intervention until a series of data processing is completed. Furthermore, if all data is not processed, this means that an error occurred in the last executed job, so examine the output, take countermeasures for the error, and restart the process.

Furthermore, even if a certain job is being executed, the results of the executed job can be sent to the job output queue 9 at any time.
You can take it out and check it.

The above is an example of a job execution management system that is an embodiment of the present invention, but the present invention is not limited to the above-mentioned embodiment, and other application examples can be considered.

For example, if an error occurs in a certain job, it is necessary to know in which data processing within the data sequence the error occurred in order to take countermeasures. Therefore, one or more pieces of data are reprocessed starting from the first piece of data in the data sequence where the error occurred. In this way, since even the data in which the error occurred is reprocessed, it is possible to specify the location where the error has occurred.

[0047] Furthermore, by accumulating and editing the job result management table, it may be useful for managing the progress of data processing work.

[0048] The invention made by the present inventor has been specifically explained based on examples, but the present invention is not limited to the above-mentioned examples, and can be modified in various ways without departing from the gist thereof. Needless to say.

[0049]

[Effects of the Invention] According to the present invention, it is possible to automatically check the results of a job that processes a continuous series of data and execute the job, so once a job is input, processing can be performed without human intervention. be able to.

Furthermore, by changing the amount of data processed per job, it is possible to automatically perform error analysis at the initial stage of processing and improve processing efficiency when processing is stable.

As a result, it becomes possible to automatically and efficiently perform data processing by sharing multiple jobs.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a computer system in which a job execution management method of the present invention is implemented.

FIG. 2 is a diagram showing an example of a job result management table used in a job execution management method according to an embodiment of the present invention.

FIG. 3 is a flowchart showing an example of the operation of a job execution management system according to an embodiment of the present invention.

FIG. 4 is a flowchart showing an example of the operation of a job execution management system according to an embodiment of the present invention.

[Explanation of symbols]

1. Calculator 2 Processing parameters 3 Job startup control section 4 Job input queue 5 Job execution section 6 Processing data 7 Job result judgment section 8 Job result management table 9 Job output queue 10 Output list

Claims (1)

[Claims] Claim 1: In a computer system that divides a desired amount of continuous data and processes each part by dividing it into several jobs, a job that processes the data is automatically started, and each of the jobs is executed. Determine the success or failure of data processing in the job based on the result, record the execution result and success or failure of the job as a data processing result, and perform data processing in a subsequent job according to the data processing result of the preceding job. If the success or failure of the preceding job indicates that data processing is possible, the job execution management method is characterized in that the amount of data to be processed in a subsequent job is increased or decreased based on the data processing result.