JPH06282446A - Job schedule generating method - Google Patents

Abstract

PURPOSE:To prevent jobs from conflicting with each other when securing a resource during execution and to reduce the consumption of a memory at the time of schedule generation. CONSTITUTION:An in-use file storage device 20 stores information on files used by the respective jobs. A job storage device 30 stores information regarding the respective jobs. An in-use resource storage device 40 stores information on resources that the respective jobs use. A resource storage device 50 stores information regarding the respective resources. An exclusive relation generation part 11 generates exclusive jobs of the respective jobs from the information in the in-use file storage device 20 and stores the job numbers of the exclusive jobs in an exclusive relation storage device 60. A schedule generation part 12 generates a schedule by repeating a process wherein the jobs are taken out of the job storage device 40 to generate a candidate job list and assignable jobs are selected by referring to the exclusive relation storage device 609 and in-use resource storage device 40 and stored in an assigned job storage part 14 and a schedule result storage device 70.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically creating a job schedule in a computer system.

[0002]

2. Description of the Related Art In creating a job schedule for a conventional computer system, the start date, the start day of the week, the context of the job,
A job described by a start time and the like is sorted by day according to a start date, a start day of the week, and the like, and then a daily job schedule is created according to the context of the job, the start time, and the like. As a publicly known example related to this, one in which a weekly operation information file is created and jobs are arranged in order of start time to create a schedule by using this is disclosed in Japanese Patent Laid-Open No. 2-216509. It is described in the official gazette.

Regarding the exclusive control of jobs by the resources used, conventionally, so-called exclusive control at the time of execution is performed in which execution of the job is started when all the resources required for the job become available. . As a publicly known example related to this, Japanese Patent Laid-Open No. 4-17057 discloses a job schedule system for performing exclusive control of shared / non-shared files between systems during execution in a complex system.

[0004]

In the above-described conventional computer system schedule creation method, when the resources necessary for the job cannot be secured at the time of executing the job, the execution of the job is postponed, resulting in the time In addition, there is a drawback that it may not be possible to finish all jobs of the day. In such a case, there is a problem that even if the resource exclusive control is performed at the time of execution like the above-described conventional resource exclusive control, it cannot be executed as scheduled in advance. Further, in order to store the resource state for performing the exclusive control, when there are a large number of resource names such as files, there is a problem that a large amount of memory is required just by managing the names.

An object of the present invention is to prevent resource contention between jobs at the time of execution by performing exclusive control by resources when creating a schedule.

Another object of the present invention is to reduce the amount of memory used by eliminating the need to manage individual resource names at the time of scheduling for shared / exclusive resources.

[0007]

The present invention provides means for storing the state of resources used by jobs, and means for storing resources used by each job and their sharing / exclusion.
By assigning jobs only when resources used by the jobs are available,
This is to prevent competition of resources between jobs at the time of execution.

Further, according to the present invention, with respect to a resource with sharing / exclusion, a means for rewriting and storing the resource information used by a job in an exclusive relationship between jobs is provided, and the exclusive relationship between jobs makes it possible to store a job depending on the resource used. By performing exclusive control, the memory for storing the resource state is reduced.

[0009]

[Operation] By performing exclusive control of resources at the time of creating a schedule, it is possible to eliminate jobs that cannot be secured on the schedule at the time of execution and cannot be executed according to the schedule, and to end daily computer operations as scheduled. Can be done. Further, by generating and storing the exclusive relationship of jobs by resources with sharing / exclusion, it is not necessary to manage the names of a large number of resources on the memory when creating a schedule.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a schedule creating system showing an embodiment of the present invention. This system is
Schedule creation device 10, used file storage device 2
0, job storage device 30, used resource storage device 40,
It comprises a resource storage device 50, an exclusive relation storage device 60, and a schedule result storage device 70. The schedule creating device 10 includes an exclusive relationship creating unit 11 and a schedule creating unit 1.
2, resource status storage unit 13, assigned job storage unit 1
It consists of 4.

As shown in FIG. 2, the used file storage device 20 stores information on files used by each job. For example, the job of job number 1 indicates that file1 is used in the exclusive mode.

In the job storage device 30, as shown in FIG.
Information about each job is stored. For example, job number 1 is composed of information such as job name A, processing time of 15 minutes, no preceding job number, no designation of start time, and the like.

As shown in FIG. 4, the used resource storage device 40 stores information on resources used by each job. For example, the job with job number 1 indicates that the resource aaa is 100 cylinders and the resource ddd is 4 units. As shown in FIG. 5, the resources aaa, bbb, and ccc are magnetic disk devices (DISK), and the resources ddd and ee are magnetic tape devices (MT).

As shown in FIG. 5, the resource storage device 50 stores information about each resource. For example, the resource name aaa is composed of information such that the resource type is DISK and the capacity is 1000.

The exclusive relationship generator 11 generates an exclusive job for each job from the information stored in the used file storage device 20. For example, in FIG. 2, the job with the job number 1 exclusively uses the file1 and the job with the job number 2 shares the file1.
Jobs are mutually exclusive jobs. Further, since the job of job number 3 uses the file 1 in common, the job of job number 2 and the job of job number 3 are not mutually exclusive. The exclusive relationship generating unit 11 stores the exclusive relationship between jobs thus generated in the exclusive relationship storage device 60.

FIG. 6 shows an example of the configuration of the exclusive relation storage device 60, in which the job numbers of the exclusive jobs are stored in association with the job numbers of the respective jobs. In FIG. 6, for example, the exclusive job number of the job of job number 1 is 2, 3,
4 is shown.

FIG. 7 is a flow chart showing the overall processing flow of the schedule creating section 12. Below, FIG.
The overall processing procedure in the schedule creating unit 12 will be described in accordance with.

The schedule creating section 12 first sets a start time and an end time (steps 701 and 70).
2). A job is assigned from the start time to the end time set here. Next, the number of initiators is set (step 703). The initiator executes a job, and the computer can execute the jobs in parallel by the number of the initiators.

Next, the resource state storage unit 13 is initialized based on each resource information of the resource storage device 50 (step 704). The resource status storage unit 13 stores the usage status of each resource. (A) of FIG.
Shows the initial state of the resource state storage unit 13 based on the example of the resource storage device 50 of FIG.
The resource name aaa is made up of information such that the resource type is DISK and the remaining capacity is 1000 cylinders.

Next, the assigned job storage unit 14 is initialized (step 705). Allocation job storage unit 14
Is composed of an initiator number, an assigned job number for the initiator number, and an end time thereof for each initiator. FIG. 11A shows an example of the initial state of the assigned job storage unit 14 when the number of initiators is 5, and there is no assigned job number and end time for all the initiator numbers.

Steps 701 to 705 are pre-processes of the schedule process, and when this is completed, the schedule process after step 6 is executed.

In the schedule process, first, the start time is set to the current time to be scheduled (step 706). Then, it is checked whether or not there is an unassigned job (step 707). If there is no unassigned job, the scheduling process ends, and if there is an unassigned job, the process proceeds to step 708. In step 708, it is checked whether or not the current time is over the end time, and if it is over, the schedule process is ended, and if it is not over, the process proceeds to step 709. In step 709, it is checked whether or not there is an initiator for which there is no assigned job in the assigned job storage unit 14 (NO in step 709), the current time is updated (step 712), and the process returns to step 707. on the other hand,
If there is an initiator for which there is no assigned job in the assigned job storage unit 14, an assigned job determination process is performed (step 710). The result of the allocation job determination process is determined (step 711), and if there is an allocation job, the process directly returns to step 707. If there is no allocation job, the current time is updated (step 712) and then step 707. Return to.

FIG. 8 is a detailed flowchart of the assigned job determination processing portion of step 710 in the flowchart of FIG.

In the allocation job determination process, first, a list of job numbers of jobs that have no preceding jobs in the candidate job list or jobs that have all completed preceding jobs at the current time is set (step 801). Figure 3
If the example of the job storage device 30 is used, the candidate job list is (1, 2, 4).

Next, one of the jobs in the candidate job list 1
On the other hand, it is checked whether there is any unchecked job in the candidate job list (step 802). If there is an unchecked job, one job is taken out from the candidate job list (step 803), the information of the job is taken out from the job storage device 30 (step 804), and step 8
Go to 05.

In step 805, it is checked whether the current time has passed the start time of the job, and if not,
Delete the job from the candidate job list (step 8
10) and returns to step 802. On the other hand, if the current time has passed the start time of the job, the process proceeds to step 806, and the exclusive job of the job is fetched from the exclusion relation storage device 60. Then, it is checked whether there is an exclusive job of the job in the allocated job storage unit 14 (step 807),
If there is, the job is deleted from the candidate job list (step 810) and the process returns to step 802. If there is no exclusive job of the job in the allocated job storage unit 14, the process proceeds to step 808, and the used resource of the job is retrieved from the used resource storage device 40. Then, the resource status storage unit 13 is checked (step 809), and if the remaining capacity of the resource used by the job is not sufficient, the job is deleted from the candidate job list (step 810) and the process returns to step 802. If the remaining capacity is sufficient, the process directly returns to step 802.

The specific processing from step 802 to step 809 will be described with reference to the candidate job list (1, 2, 4) according to the example of the job storage device 30 of FIG. 3 and the example of the used resource storage device 40 of FIG. An example of the exclusive relationship storage device 60 of FIG. 6, an example of the resource state storage unit 13 of FIG. 10A, and an example of the allocation job storage unit 14 of FIG. 11A will be described.

The job of job number 1 is taken out from the candidate job list (step 803), and the job storage device 30
The job information of job number 1 is retrieved from (step 8
04), the determination in step 805 is performed. Here, assuming that the current time is 9:00, the start time is not specified for the job with job number 1, so the process proceeds to step 806, and the exclusive jobs 2, 3 with the job number 1 are stored in the exclusive relationship storage device 60.
4 is taken out and the judgment of step 807 is performed. From FIG. 11A, 2, 3, 4 are stored in the assigned job storage unit 14.
Since there is no job of, the process proceeds to step 808,
The used resource information of the job number 1 is retrieved from the used resource storage device 40, and the determination in step 809 is performed. From FIG. 10A, the remaining capacity of the resource aaa is 100.
Since 0 cylinders and the remaining capacity of resource ddd are 10 units, which are all larger than the resource usage amount of job number 1,
The job number 1 is left in the candidate job list and the process returns to step 802. Similarly, when the processing of the jobs with job numbers 2 and 4 is advanced, the start time of the job with job number 2 is 9:12, which is after the current time of 9:00, so the job with job number 2 is a candidate job. Deleted from the list, the candidate job list becomes (1,4).

Returning to the explanation of the flow chart of FIG. 8, if it is determined in step 802 that there are no unchecked jobs in the candidate job list, the process advances to step 811 to check whether the candidate job list is empty. Here, if the candidate job list is empty, the allocation job determination process ends. If the candidate job list is not empty, one assigned job is selected from the candidate job list (step 812) and the information of the job is retrieved from the job storage device 30 (step 813). Then, the assigned job number and the end time are set to one of the initiators having no assigned job in the assigned job storage unit 14 (step 814). Next, the used resource of the assigned job is retrieved from the used resource storage device 40 (step 815), and the remaining capacity of the resource name corresponding to the used resource of the assigned job of the resource status storage unit 13 is subtracted (step 816). Finally, the assigned job is stored in the schedule result storage device 70 (step 817).

Specific processing of steps 812 to 817 will be described below with reference to the aforementioned example.

From the candidate job list (1, 4), the job of job number 1 is selected as the allocation job (step 812), and the information of the job of job number 1 is retrieved from the job storage device 30 (step 813). Next, FIG.
1 (a), the assigned job number of the initiator number 1 in the assigned job storage unit 14 is set to 1, and the end time is obtained by adding 15 minutes of the processing time of the job number 1 to the current time of 9:00. Is set (step 814).
As a result, the allocated job storage unit 14 is shown in FIG.
It becomes the state of.

Next, the used resource of job number 1 is taken out from the used resource storage device 50 (step 81).
5), 900 cylinders obtained by subtracting the usage amount 100 cylinders from 1000 cylinders are set to the remaining capacity of the resource name aaa in the resource state storage unit 13 of FIG. 10A, and the remaining capacity of the resource name ddd is set to 10 units. Then, 6 units are set by subtracting 4 units from (Step 816). As a result, the resource state storage unit 13 becomes the state of (b) of FIG.

Finally, the assigned job is stored in the job number 1 of the schedule result storage device 70 (step 817). FIG. 12 is an example of the schedule result storage device 70. For example, the job with the job number 1 which is the currently assigned job has information such that the execution start time is 9:00 and the execution end time is 9:15.

FIG. 9 is a detailed flowchart of the current time update processing portion of step 712 in the flowchart of FIG.

First, the current time is advanced (step 9
01), it is checked whether there is a job (finished job) whose end time is earlier than the current time in the allocated job storage unit 14 (step 902).
On the other hand, if there is an end job, the used resource storage device 40
The resource used by the end job is retrieved from (step 9)
03). Then, in the remaining capacity of the resource name corresponding to the used resource of the end job in the resource status storage unit 13,
The usage amount of the end job is added (step 904), and the end job is removed from the assigned job storage unit 14 (step 905).

In the above example, the current time is 9:00.
When the resource status storage unit 13 is in the state of FIG. 10B and the assigned job storage unit 14 is in the state of FIG. 11B, if the current time is updated to 9:15, The job is an end job. At this time, in the current time update processing, the used resource of job number 1 is taken out from the used resource storage device 40 (step 903), and the usage amount of 100 cylinders is used for the cylinder of the remaining capacity 900 of the resource name aaa in FIG. The remaining capacity is set to 1000 cylinders obtained by adding 4 to the remaining capacity 6 of the resource name ddd, and the remaining capacity is set to 10 units (step 9).
04). The resource state storage unit 13 is now in the state shown in FIG. Also, as shown in FIG. 11B, the assigned job number 1 of the initiator number 1 and the end time 9: 1
No. 5 is set (step 905). As a result, the assigned job storage unit 14 also enters the state shown in FIG.

[0038]

According to the first aspect of the invention, when creating a job schedule, exclusive control of jobs by resources is performed at the time of creating a schedule. Therefore, there are no jobs that cannot be secured as resources and cannot be executed as scheduled. The operation of the computer can be finished as scheduled.

According to the second aspect of the invention, in the case of a resource with sharing / exclusion, an exclusive relationship between jobs is generated and stored in advance, so that individual resource names are not managed at the time of scheduling. Therefore, the memory usage is small.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a schedule creating system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a specific example of a used file for each job in a used file storage device.

FIG. 3 is a diagram showing a specific example of job information in a job storage device.

FIG. 4 is a diagram showing a specific example of used resources for each job in a used resource storage device.

FIG. 5 is a diagram showing a specific example of resource information of a resource storage device.

FIG. 6 is a diagram showing a specific example of an exclusive relationship between jobs in an exclusive relationship storage device.

FIG. 7 is an overall flowchart of processing of a schedule creating unit in the schedule creating apparatus.

FIG. 8 is a detailed flowchart of an assigned job determination process in FIG.

9 is a detailed flowchart of a current time update process in FIG.

FIG. 10 is a diagram showing a specific example of a resource status storage unit of the schedule creation device.

FIG. 11 is a diagram showing a specific example of an assigned job storage unit of the schedule creation device.

FIG. 12 is a diagram showing a specific example of a schedule result storage device.

[Explanation of symbols]

 10 Schedule Creation Device 11 Exclusive Relationship Generation Unit 12 Schedule Creation Unit 13 Resource State Storage Unit 14 Assigned Job Storage Unit 20 Used File Storage Device 30 Job Storage Device 40 Used Resource Storage Device 50 Resource Storage Device 60 Exclusive Relationship Storage Device 70 Schedule Result Storage apparatus

Claims (2)

[Claims] 1. A method for automatically creating a job schedule in a computer system, comprising means for storing the state of resources used by jobs, and means for storing resources used by each job and their sharing / exclusion. A job schedule creating method characterized in that when a job is assigned, the assignment is performed only when the resource used by the job is in an available state. 2. The job schedule creation method according to claim 1, further comprising means for rewriting and storing resource information used by jobs for resources with sharing / exclusion in an exclusive relationship between jobs, and A job schedule creation method characterized by performing exclusive control of jobs by resources used in an exclusive relationship.