JP3664079B2 - Job stopping method and apparatus in system freeze - Google Patents

Description

TECHNICAL FIELD The present invention relates to a job execution control method in system freeze, and more particularly, to a job stop method and apparatus for the purpose of shortening system freeze time and job restart time.
BACKGROUND ART In general, in a computer system, in order to execute a job without being aware of the operation time of the system, job execution information is stored in an external storage device when the system is stopped, and when the system is restarted, the stored information is stored. Originally, it has a function of recovering the job state to the stored time before restart and restarting job execution. The function of storing the job execution information in the external storage device is called system freeze. When the system is restarted, the job execution is restarted from the system freeze state based on the information stored in the system freeze. This is called job restart.
Conventionally, in system freeze / restart, among jobs that were executed when the system was stopped, it was difficult to resume execution of jobs that were accessing physical devices such as tape devices after job restart. I was excluded from the system freeze.
In addition, the conventional system freeze is not only performed when a system stop request is received from the user, but also has a function to restart the job after power recovery by performing system freeze using an emergency battery in the event of a power failure during system operation. I have. This system freeze at the time of a power failure is called a power failure freeze. However, the power outage freeze processing time cannot exceed the duration of the emergency battery.
In the conventional technology, jobs that cannot be restarted are stopped among the jobs that are being executed when the system is stopped, and the remaining jobs are subject to freezing. Among the remaining jobs, there are jobs that have secured a large amount of memory. If so, there is a problem that it takes time to freeze the system in which the job is stored in the external storage device. In particular, in a power outage freeze, there is a limitation that the system freeze time cannot exceed the duration of the emergency battery, so it is desirable that the power outage freeze time be as short as possible. Further, when the system freeze time of the job that secures a large amount of memory is compared with the time until execution of the job and the memory securing process is completed, the latter is shorter.
A main object of the present invention is to reduce the time required for system freeze and job restart by reducing job execution information on the main storage device stored in the external storage device in system freeze / restart. .
DISCLOSURE OF THE INVENTION Means of the invention for stopping a job whose CPU time used from the start of job execution to before system freeze is less than a reference time will be described. When a system stop occurs on an arbitrary computer that executes multiple jobs, after a job that cannot be restarted due to system freeze is stopped, from among the jobs that can be restarted, from the start of job execution to before system freeze That is, a job whose CPU time used is less than the reference time is stopped to be excluded from the system freeze target.
As a result, the time required for system freeze is shorter than the time required for system freeze of a job whose CPU time does not satisfy the standard, compared to when only a job that cannot be restarted is stopped. This is particularly effective for jobs that have secured a large amount of memory. Furthermore, by stopping a job whose CPU time does not satisfy the standard, it is not necessary to restart the job, and the processing time is reduced accordingly.
An invention means for stopping a job in which the amount of memory secured by the job exceeds a reference amount when the system freeze is started will be described. When a system stop occurs on an arbitrary computer that executes multiple jobs, because the system freezes, after stopping a job that cannot be restarted, among the jobs that can be restarted, the amount of memory secured by the job is By stopping jobs that exceed the reference amount, they are removed from the system freeze target.
As a result, the time required for system freeze is shorter than the time required for system freeze of a job in which the amount of memory secured by the job exceeds the reference amount compared to when only a job that cannot be restarted is stopped. . Furthermore, by stopping a job in which the secured memory amount exceeds the reference amount, it is not necessary to restart the job, and the processing time is reduced accordingly. An invention means for stopping a job whose job class is equal to or lower than the reference job class when starting the system freeze will be described. When a system stop occurs on an arbitrary computer that executes multiple jobs, because of system freeze, after stopping a job that cannot be restarted, among the jobs that can be restarted, the job class of the job is the reference By stopping jobs below the job class, they are removed from the system freeze target.
As a result, the time required for the system freeze is shortened by the time required for the system freeze of the job whose job class is equal to or lower than the reference job class, compared to the case where only the job that cannot be restarted is stopped. Furthermore, by stopping jobs whose job class is equal to or lower than the reference job class, it is not necessary to restart the job, and the processing time corresponding to that job is shortened.
[Brief description of the drawings]
FIG. 1 is a block diagram of a system configuration showing an embodiment of the present invention, FIG. 2 is a flowchart showing a first embodiment of job stop processing in the system freeze of the present invention, and FIG. FIG. 4 is a flowchart showing a third embodiment of the job stop process in the system freeze according to the present invention. FIG. 4 is a flowchart showing the third embodiment of the job stop process in the system freeze according to the present invention.
BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of system freeze / restart to which the present invention is applied. The computer 10 includes a CPU, a memory, and the like (not shown), and is connected to an external storage device 20. The computer 10 includes a system monitoring unit 11, a freeze / restart control unit 12, and a job stop control unit 13.
First, general execution control of system freeze / restart in such a computer system will be described with reference to FIG.
When there is a system stop request due to freeze from the user, the system monitoring unit 11 notifies the freeze / restart control unit 13 to that effect. Further, not only the system stop request from the user, but also the freeze / restart control unit 13 is notified in the case of a power failure freeze at the time of power failure occurrence. Further, when a restart request is received from the user, the fact is notified to the freeze / restart control unit 13.
When there is a freeze request from the system monitoring unit 11, the freeze / restart control unit 12 determines whether the job being executed on the computer can be restarted. If the restart is impossible, the job stop unit is notified of the job stop. If restart is possible, information (job execution information 21) necessary for restarting the job, such as job register values, stack information, and data information, is stored in the external storage device 20. The opportunity of system freeze may be given by the user or by power failure freeze. When restarting a job in response to an instruction from the system monitoring unit 11, the freeze / restart control unit 13 reads job execution information 21 corresponding to the job stored in the external storage device 20 of the computer, and The job is restored from the information to the state at the time of the system freeze, and the processing is resumed.
The job stop control unit 13 stops the instructed job. The present invention relates to the execution control of the subsequent system freeze.
Next, an embodiment of the system freeze control according to the present invention will be described with reference to FIG. 1 and FIG. 2 to FIG.
FIG. 2 is a flowchart showing a first embodiment of job stop processing in the system freeze according to the present invention. When there is a system stop request from the user or due to a power outage freeze, the system monitoring unit 11 receives the system stop request (201). The freeze / restart control unit 12 determines a job that cannot be restarted from the jobs that were being executed when the stop request was received, and notifies the job stop unit 13 of the job that cannot be restarted. The control unit 13 stops the job (202). Next, the freeze / restart control unit 12 determines the CPU time used from the start of job execution to before the system freeze among the remaining jobs, and satisfies the reference time. The job stop control unit 13 notifies the job stop control unit 13 of the absence, and the job stop control unit 13 stops the job (203). Thereafter, the freeze / restart control unit 12 sets the remaining jobs as system freeze targets and performs system freeze (204). When the system freeze of all jobs is completed, the system is stopped (205).
FIG. 3 is a flowchart showing a second embodiment of job stop processing in the system freeze according to the present invention. When there is a system stop request from the user or due to a power outage freeze, the system monitoring unit 11 accepts the system stop request (301). The freeze / restart control unit 12 determines a job that cannot be restarted from the jobs that were being executed when the stop request was received, and notifies the job stop unit 13 of the job that cannot be restarted. The control unit 13 stops the job (302). Next, the freeze / restart control unit 12 determines which of the remaining jobs the memory amount secured by the job exceeds the reference amount, and determines the job exceeding the reference memory amount as the job Notifying the stop control unit 13, the job stop control unit 13 stops the job (303). Thereafter, the freeze / restart control unit 12 sets the remaining jobs as system freeze targets and performs system freeze (304). When the system freeze of all jobs is completed, the system is stopped (305).
FIG. 4 is a flowchart showing a third embodiment of job stop processing in the system freeze according to the present invention. When there is a system stop request from the user or due to a power outage freeze, the system monitoring unit 11 accepts the system stop request (401). The freeze / restart control unit 12 determines a job that cannot be restarted from among the jobs that were being executed when the stop request was received, notifies the job stop unit 13 of the job that cannot be restarted, and stops the job The control unit 13 stops the job (402). Next, the freeze / restart control unit 12 determines whether or not the job class of the job is equal to or lower than the reference job class among the remaining jobs, and the job stop control unit 13 determines whether the job class is equal to or lower than the reference job class. The job stop control unit 13 stops the job (403). Thereafter, the freeze / restart control unit 12 sets the remaining jobs as system freeze targets and freezes the system (404). When the system freeze of all jobs is completed, the system is stopped (405).
Since the time required for the system freeze / restart of the present invention depends on the reference values of job attributes (CPU usage time, reserved memory amount, job class), the reference value is determined so as to be optimal for the system.
Industrial Applicability As described above, according to the present invention, the system freeze / restart execution time can be shortened by stopping the job from the job attribute value in the system freeze / restart process. Can do. In particular, the present invention has a great effect in a power failure freeze in which the system freeze time is limited.

Claims (6)

In a method in which a computer executing multiple jobs stops the job,
When performing the system freezes, it determines whether the freeze restart controller saves the job execution information for restart each running job based on criteria comprising a job execution record, freeze Li A job stopping method for storing job execution information for restarting a job that satisfies a criterion for shortening a processing time at the start in an external storage device. In the job stopping process according to claim 1, as a criterion for determining whether to save the job execution information for restart each job in the freeze restart controller is running, start the system freezes A job stopping method , wherein a job whose memory amount reserved by the job is smaller than the reference amount is selected as a system freeze target . In the job stopping process according to claim 1, as a criterion for determining whether to save the job execution information for restart each job in the freeze restart controller is running, start the system freezes A job stopping method , wherein a job whose CPU time used by the job is longer than the reference time is selected as a system freeze target . In the job stopping device of a computer that executes a plurality of jobs,
When performing the system freezes, determines freeze restart controller based on whether the criteria consisting of job execution record stores the job execution information for each of the job restart running and the freeze A job stop device comprising: an external storage device for storing job execution information for restarting a job that satisfies a criterion for reducing a processing time at freeze / restart by the restart control unit . In the job stopping device according to claim 4, as a criterion for determining whether to save the job execution information for restart each job in the freeze restart controller is running, start the system freezes job stopping device, characterized in that the amount of memory the job was secured when to select a job less than the amount of the reference as an object of the system freezes. In the job stopping device according to claim 4, as a criterion for determining whether to save the job execution information for restart each job in the freeze restart controller is running, start the system freezes job stopping device, characterized in that the CPU time the job was used to selects a longer job than the time of the reference as an object of the system freezes.

Images