JPH0283655A - Job schedule stopping system - Google Patents

Abstract

PURPOSE:To easily define a system by affecting correspondence defined between plural resources in a local host and a local job class whose schedule should be stopped against the fault of the resource on the schedule of a global job. CONSTITUTION:A schedule stoppage request means 28 includes a means to inform the name of a job class relating to faulty resource to a global host 10 via an inter-host communication means 40. The global host 10 includes a global job schedule stopping means 14 as a second setting means which sets control information to stop the schedule for the informing origin host(local host 20) of a corresponding job class on a global job class definition table 12 based on the output information of the informing means. In such a way, it is not required to define the correspondence with the global job whose schedule should be stopped against the resource in the local host 20 and the fault of the resource in the global host 10, which facilitate the definition of the system.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention is utilized in a job schedule stopping method for a multi-host system. In particular, it relates to a method for stopping job schedules between hosts when a resource failure occurs in a local host.

〗〗　　　　

In the job schedule stop method, the present invention reflects the correspondence defined between multiple resources in the local host and the local job class whose schedule should be stopped in response to a failure in the global job schedule. This simplifies system definition because there is no need to define on the global host the association between resources in the local host and global job classes whose schedules should be stopped in response to a failure.

[Conventional technology]

A multihost system consists of one global host that handles the human resources and scheduling of global jobs, and one or more local hosts that provide the human resources for local jobs and the execution of local and global jobs. Global hosts also often overlap local hosts.

When some resources become unusable on a local host due to a failure, it is necessary to stop the schedule of jobs that use those resources, regardless of whether they are global jobs or local jobs.

Which job to stop is determined based on the job class declared for each job.

Conventionally, job scheduling methods determine the job class whose schedule should be stopped due to a failure on both the global host and the local host, and then stop the schedule for that job class.
Execution of the job that uses that resource was stopped.

[Problem that the invention seeks to solve]

However, in this conventional job scheduling two-stop method, global jobs are scheduled on the global host and local jobs are scheduled on the local host, so it is difficult to associate the faulty resource with the job class whose schedule should be stopped. This had to be done on both the global host and the local host, which had the disadvantage that the system definition was complicated.

The present invention solves the above-mentioned drawbacks, and eliminates the need to define on the global host the correspondence between resources in the local host and job classes whose schedules should be stopped in response to a failure of the resources, making system definition of jobs easy. The purpose is to provide a schedule stopping method.

[Means for solving problems]

The present invention provides a first definition table that stores control information for a job class salt declared for a job executed on a local host, and executes a job based on the contents of this first definition table. a local host that includes a first scheduling means for requesting and a first setting means for setting control information for stopping the schedule of a job class related to the resource in the first definition table in the event of a resource failure in the local host; , a second definition table that stores control information for the job class salt declared for jobs executed on other hosts, and a job is sent to the corresponding local host based on the contents of this second definition table. and a global host including a second scheduling means for requesting execution, wherein the first setting means includes means for notifying the global host of a job class salt associated with the faulty resource; The global host is characterized in that it includes a second setting means for setting, in the second definition table, control information for stopping the schedule for the notification source host of the corresponding job class based on the output information of the notification means. shall be.

[Effect]

On the local host, when a fault is detected in a resource within the local host, the first setting means sets control information to stop the schedule of the job class related to the faulty resource in the first definition table, and also sends the fault to the global host. Notify the job class salt associated with the resource. In the global host, based on this notification, the second setting means sets control information for stopping the schedule for the notification source local host of the corresponding job class in the second definition table.

With the above operation, it is not necessary to define on the global host the association between the resources in the local host and the global job whose schedule should be stopped in response to a failure thereof, and system definition can be made easily.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings. 1st
The figure is a block diagram of a job schedule stopping device according to an embodiment of the present invention. In FIG. 1, the job schedule stopping device includes a local job manual means 21 that manually inputs jobs to be executed on its own host, and a local job input means 21.
A local spool file 24 that stores the output of the job, a local job class definition table 22 as the first definition table that stores control information for the job class salt declared for the job, and the contents of the local job class definition table 22 A local scheduler 23 selects and requests execution of a job within a multiplicity range based on a first schedule, a job execution unit 25 executes a job based on a request from the local scheduler 23, and resources within the host itself. The failure detection means 26 serves as a first setting means for setting, in the local job class definition table 22, control information for stopping the schedule of a job class that uses that resource in the event of a failure. ．． 262, a local host 20 including a resource class correspondence table 27 and a schedule stop requesting means 28, a global job manual means 11 for inputting jobs to be executed on other hosts, and a control for the job class salt declared for the job. As a second definition table in which information is stored, jobs are selected within the multiplicity range based on the contents of the global job definition table 12 and the global job definition table 12, and the corresponding local host (local host 20) is selected.
A global host 10 including a global job scheduling means 13 as a second scheduling means for requesting execution, and a shared spool file in which the output of the global job manual means 11 is stored and the stored job is executed by the job execution means 25. 30, and inter-host communication means 40 for notifying the job execution means 25 of the request from the global job scheduling means 13.

Here, the feature of the present invention is that the schedule stop requesting means 28 transmits the job class salt related to the faulty resource to the global host 1 via the inter-host communication means 40.
The global host 10 sends control information for stopping the schedule for the notification source host (local host 20) of the corresponding job class based on the output information of the notification means to the global job class definition table 12. A global job schedule stopping means 14 is included as a second setting means for setting.

Here, there may be a plurality of local hosts, and a global host may also serve as a local host.

The operation of the job schedule stopping device having such a configuration will be explained.

Table 1 is a global job class definition table of the job schedule stopping device of the present invention, in which job class salt C,
It consists of a multiplicity m for each host constituting the multi-host system and a schedule stop flag for each host.

Table 2 is a local job class definition table of the job schedule stop device of the present invention, and is composed of a job class salt C, multiplicity n, and schedule stop flags corresponding to each resource in the local host. .

Table 3 is a resource class correspondence table of the job schedule stopping device of the present invention, which shows job class C and resources A and B used by jobs of that job class.

. . . This is a correspondence table.

(Hereinafter referred to as the margin of this page) FIG. 2 is a flowchart showing the operation of the job schedule stopping device of the present invention.

In FIGS. 1 and 2, global job scheduling is performed as follows.

In the global host 10, the global job input by the global job manual means 11 is registered in the shared spool file 30 (Sl).

This global job declares a global job class name that represents the attributes of the job, and a local host name that represents which local host the job should be executed on.

The global job scheduling means 13 refers to the contents of the global job class definition table 12.

This global job class definition table 12 shows, for each global job class, the number of jobs that can be executed simultaneously (multiplicity) for each local host. For example, host 1 of global job class C1
The multiplicity for is m11.

The global job scheduling means 13 selects a global job from the shared spool file within this multiplicity range and sends the job to the job execution means 25 of the local host 20 that is to execute the job via the inter-host communication means 30. (S2).

Furthermore, local job scheduling is performed as follows. In the local host 20, the local job manually created by the local job manual means 11 is registered in the local spool file 24 (Sl
l). For this local job, a local job class salt representing the attributes of the job is declared.

The local job scheduling means 23 refers to the local job class definition table 22. This local job class definition table 22 shows the number of jobs that can be executed simultaneously (multiplicity) for each local job class. For example, the multiplicity of local job class C1 is n.

The local job scheduling means 23 requests the job execution means 25 to select a local job from the local job file 24 within this multiplicity range and execute the job (S12).

The job execution unit 25 executes a local job manually executed from the own local host 20 and a global job manually executed from the global host (S13). When all jobs are completed (314), the operation ends.

If a failure occurs in a resource on the local host, stop the job schedule by following the steps below.

The failure of the resource is detected by the failure detection means 26 (315). When the failure detection means 26 detects a failure,
The resource name is notified to the schedule stop requesting means 28 (S16).

The schedule stop requesting means 28 uses the resource class correspondence table 2.
7, find out the local job class of the job that uses that resource, and turn on the schedule stop flag corresponding to the resource that caused the failure in the corresponding local job class entry in the local job class definition table ( 317).

The local job scheduling means does not request execution of local jobs belonging to a local job class for which at least one schedule stop flag is "on". This will stop the scheduling of local jobs related to the failure (31
B).

At the same time, the schedule stop request means 28 transmits the job class name to the global job schedule stop means 1 in the global host 10 via the inter-host communication means 30.
4 (S19). The global job schedule stopping means 14 is configured to stop the notification source local host 2 of the job class notified from the global job class definition table.
The schedule stop flag corresponding to 0 is turned on (S3, S4).

For a global job class whose schedule stop flag is "on," the global job scheduling unit 13 does not request execution of a global job that belongs to that global job class and is requested to be executed by the corresponding local host. As a result, the schedule of the global job related to the failure is stopped (S5).

〔Effect of the invention〕

As described above, the present invention has an excellent effect of simplifying system definition since there is no need to define on the global host the association between a resource failure in a local host and a job class whose schedule should be stopped.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a job schedule stopping device according to an embodiment of the present invention. FIG. 2 is a flowchart showing the operation of the job schedule stopping device of the present invention. DESCRIPTION OF SYMBOLS 10... Global host, 11... Global job manual means, 12... Global job class definition table, 13... Global job schedule means, 1
4...Global job schedule stopping means, 20
. . . local host, 21 . . . local job input means, 22 . . . local job class definition table, 23.
. . . Local job scheduling means, 24 . . . Local job execution means, 2
6. , 26.2... Failure detection means, 27... Resource class correspondence table, 28... Schedule stop requesting means,
30... Inter-host communication means, 4o... Shared spool file.

Claims (1)

[Claims] 1. A first definition table storing control information for job class names declared for jobs executed on the local host, based on the contents of this first definition table. a first scheduling means for requesting execution of a job; and a first setting means for setting in the first definition table control information for stopping the schedule of a job class related to the resource in the event of a failure of a resource within the host; a second definition table that stores control information for job class names declared for jobs executed on other hosts; In a job schedule stop method comprising a global host including a second scheduling means for requesting a host to execute a job, the first setting means notifies the global host of a job class name related to the faulty resource. The global host includes a second setting means for setting, in the second definition table, control information for stopping the schedule for the notification source host of the corresponding job class based on the output information of the notification means. A job schedule stop method characterized by comprising: