JPH07253893A - System for managing task priority order - Google Patents

Abstract

PURPOSE:To prevent the excess execution of a task of high priority and to improve the uniformity of processing by making an adjustment of the priority of each task based on its execution time. CONSTITUTION:A storage part with plural tasks each with a priority is connected to an operation system with an internal timer. Each task in the storage part executes the processing based on the priority in the computer system. Each task is generated abased on the priority and a scheduler 2 which monitors the execution time for every execution of each task compares the execution time monitored by the scheduler with the prescribed reference execution time. A priority adjustment means 3 which reduces the priority of the task than the priority of the other tasks when the execution time of the task being executed exceeds the reference execution time is provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a task priority management method used in a computer system for performing multitask processing, and in particular, by adjusting the priority of each task based on its execution time, The present invention relates to a task priority management method that can prevent excessive execution of high priority tasks.

[0002]

2. Description of the Related Art In recent years, for example, in a computer system for performing process control of various plant equipment, a task priority management system has been adopted in order to execute a plurality of tasks corresponding to their priorities.

In this kind of task priority management system, different priorities are set for the respective tasks, and the task having the highest priority among the tasks to be executed is executed. After the execution of this task is completed, the task with the next highest priority is executed, and so on. If there is an interrupt by another task during execution of the task, the priority of the task currently being executed is compared with the priority of the interrupted task, and the task with the higher priority is executed. Furthermore, the priority of such tasks can be changed individually.

[0004]

However, in the above computer system, if the currently executing task is always in operation for some reason, the task having a lower priority than the currently executing task cannot be executed. Therefore, there is a problem that the uniformity of the treatment is deteriorated.

Further, when the priority of a certain task is changed, the execution order of other tasks in the same job is reversed due to this change, and a problem occurs in the related operation (communication) between the tasks. , There may be inconsistency in the job operation.

The present invention has been made in consideration of the above-mentioned circumstances, and by adjusting the priority of each task based on its execution time, it is possible to prevent an excessive execution of a high-priority task and to make the processing uniform. It is an object of the present invention to provide a task priority management method that can improve the performance.

Another object of the present invention is to prevent the priority of another task belonging to the same job as this task from changing the priority order of the same job when changing the priority of a certain task. By changing, it is to ensure the consistency of the job.

[0008]

According to a first aspect of the invention, a storage unit having a plurality of tasks each having a priority is connected to an operation system having an internal timer, and each task in the storage unit has the priority. In a computer system that executes processing based on the priority, while generating each of the tasks based on the priority, a scheduler that monitors the execution time for each execution of these tasks, and the execution time monitored by the scheduler and A priority adjusting means for comparing predetermined reference execution times, and lowering the priority of the task at least one of the other tasks when the execution time of the currently executed task exceeds the reference execution time. It is a task priority management system with.

According to a second aspect of the invention, a storage unit having a plurality of jobs, each of which has a priority, is connected to an operating system having an internal timer, and each task in the storage unit has the priority. In a computer system that executes processing based on the above, a job information storage unit that stores a job identifier for each job and task identifiers that belong to the same job for each job identifier stored in this job information storage unit , Task information storage means in which each task priority corresponding to each of these task identifiers and other task identifiers belonging to the same job are stored, and some task priority among the task priorities stored in this task information storage means When the degree is changed, based on the stored contents of the task information storage means and the job information storage means, A task priority management method including an intra-job priority adjusting means for changing the priority of a task having a changed task priority and another task belonging to the same job so as not to change the priority order of the same job. Is.

[0010]

Therefore, according to the invention corresponding to claim 1, by taking the above-mentioned means, the scheduler generates each task based on the priority, while monitoring the execution time for each execution of each task. However, the priority adjusting means compares the execution time monitored by this scheduler with a predetermined reference execution time, and when the execution time of the currently executing task exceeds the reference execution time, the priority of the task is set. It lowers the priority of at least one other task, thus adjusting the priority of each task based on its execution time to prevent excessive execution of high-priority tasks and improve processing uniformity. Can be made.

Further, the invention corresponding to claim 2 is that the job information storage means in which the job identifier is stored for each job and each task identifier belonging to the same job for each job identifier stored in the job information storage means. The task information storage means for storing the identifiers, the respective task priorities corresponding to the respective task identifiers, and the other task identifiers belonging to the same job are provided, and the in-job priority adjustment means stores the task information storage means. When a certain task priority of the task priorities is changed, the priority of another task belonging to the same job as the task having the changed task priority is based on the stored contents of the task information storage means and the job information storage means. Since the degree is changed so as not to change the order of the priority in the same job, it is possible to ensure the consistency of the jobs.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a computer system applied to a task priority management system according to an embodiment of the present invention. In this computer system, an OS (operating system) 3 having an internal timer 1 and a scheduler 2 stores a storage unit 4 having a plurality of jobs each having a plurality of tasks each having a priority and each job in the storage unit 4. It is connected to the CPU 5 that can execute each task.

Here, the storage unit 4, as shown in FIG.
A job table (job information storage means) 6 in which a job ID (identifier) is stored for each job, task IDs belonging to the same job for each job ID stored in the job table 6, and these task IDs And a task table (task information storage means) 8 in which task information 7 including each task priority corresponding to the above and other task IDs belonging to the same job is stored.

On the other hand, the scheduler 2 has a function of generating each task based on the priority and monitoring the execution time for each execution of each task. The OS 3 compares the execution time monitored by the scheduler 2 with a predetermined reference execution time, and when the execution time of the task currently being executed exceeds the reference execution time, the priority of the task is at least one other. It has the function of lowering the priority of the task.

When a certain task priority among the task priorities stored in the task table 8 is changed, the OS 2 changes the task based on the contents stored in the task table 8 and the job table 6. It has a function to change the priority of a task having a priority and that of other tasks belonging to the same job so as not to change the order of the priority in the same job. That is, the OS 2 is configured to include the priority adjusting means and the in-job priority adjusting means.

Next, the operation of the task priority management system by the computer system thus configured will be described. It is assumed that there are two jobs to be executed, and job IDs of these jobs are j1 and j2.

The job j1 is composed of tasks A to C having task IDs A to C. The priorities of the tasks A to C are 200 for task A, 250 for task B, and 300 for task C. It should be noted that this priority corresponds to the height of the numeral, and indicates that in job j1, task C, task B, and task A are executed in this order.

Similarly, the job j2 has task IDs D and E.
Task D, E. The priorities of the tasks D and E are 160 for the task D and 180 for the task E. Each of these tasks A to E is in a waiting state.

The standard execution time set in the OS 3 is 1000 msec, and the interrupt period of the internal timer 1 is 10 msec. Now, as shown in FIG. 3, the scheduler 2 judges whether or not there is a task dispatch request (ST1), and when the result of the judgment is that there is a dispatch request, the task has the highest priority among the tasks A to E. A task C having a degree is dispatched (ST2). At this time, the scheduler 2 initializes the monitoring time for monitoring the execution time of the task C (ST3).

On the other hand, when the OS 3 receives a timer interrupt from the internal timer 1 every 10 msec as shown in FIG. 4, it determines whether or not there is a task currently being executed (ST1
1) Since the determination result indicates that task C is executed, the monitoring time is updated (ST12).

Subsequently, the OS 3 has the monitoring time updated in step ST12 of 1000 ms as the reference execution time.
It is determined whether or not ec is exceeded (ST13), and when the determination result shows no, the next timer interrupt is awaited.

Thereafter, the processes of steps ST11 to ST13 are executed every 10 msec until the monitoring time exceeds 1000 msec. Where task C is 1
It is assumed that the processing is executed over 000 msec. OS3
Determines in step ST13 that the monitoring time has exceeded 1000 msec and reduces the priority of task C currently being executed, "300", by, for example, 1/2, to "150" (ST14), Initialize the monitoring time (ST1
5).

Therefore, the priority "250" of the task B becomes the highest among the priorities of the tasks A to E, and the task B becomes executable. In this way, since the priority of each task is adjusted based on its execution time, it is possible to prevent excessive execution of high-priority tasks and improve processing uniformity.

Next, it is assumed that the scheduler 2 dispatches the task B instead of the task C. At this time, if the task B executes the processing based on the processing result of the task C, the priorities of the tasks may be reversed and the related operation between the tasks may be defective as in the conventional case. Therefore,
In order to prevent such a problem, when the task priority is changed, the OS 3 changes the priority of the entire task group in the job to which the task belongs.

For example, as shown in FIG. 5, when changing the priority of the task C as in step ST14, the OS 3 reads the job ID to which the task C belongs by referring to the task table 8 (ST21). , Read job I
Based on D, the first task ID belonging to the same job is read from the job table 6 (ST22).

Subsequently, the OS 3 sets the task information 7 in the task table 8 so as to lower the priority of this task ID.
Is updated and stored (ST23), and the task table 8
Is executed to read the next other task ID belonging to the same job (ST24), the presence or absence of the next other task ID is determined (ST25), and when there is the next other task ID, the other task ID is indicated by this other task ID. The process returns to step ST23 to lower the priority of the task.

Thereafter, the OS 3 proceeds to step ST23 until the priorities of all tasks belonging to the same job are lowered.
Through the processing of step ST25, and step ST2
If it is determined in step 5 that there is no other task ID, the priority lowering process ends.

In this way, when changing the priority of a task, all the priorities of other tasks belonging to the same job as this task are changed, so that the order of priority in the same job is prevented from being reversed. Job consistency can be ensured.

As described above, according to this embodiment, the scheduler 2 generates each task based on the priority, updates the monitoring time each time each of these tasks is executed, and updates the OS.
3 compares the monitoring time with the reference execution time, and when the monitoring time exceeds the reference execution time, the priority of the currently executing task is lowered below the priority of at least one other task. By adjusting the priority of the task based on its execution time, it is possible to execute the task of the low priority, prevent the task of the high priority from being excessively executed, and improve the processing uniformity.

Further, according to the present embodiment, when the OS 3 changes a certain task priority among the task priorities stored in the task table 8, the contents stored in the task table 8 and the job table 6 are changed. Based on this, the priority of other tasks that belong to the same job as the task with the changed task priority is changed so that the priority order of the same job is not changed. It is possible to prevent the possibility that a problem will occur, to ensure job consistency, and to balance the processing of the entire system.

In the above embodiment, the case where the OS 3 adjusts the priority has been described, but the present invention is not limited to this, and when the execution of a process that exceeds the reference execution time is detected, this detection result is used as a task or a job-related task. Even if the task side is configured to adjust the priority by notifying the group, the present invention can be similarly implemented and the same effect can be obtained.
In addition, the present invention can be modified in various ways without departing from the scope of the invention.

[0032]

As described above, according to the invention of claim 1, the scheduler generates each task based on the priority, while monitoring the execution time for each execution of each task and adjusting the priority. The means compares the execution time monitored by this scheduler with a predetermined reference execution time, and when the execution time of the task currently being executed exceeds the reference execution time,
Since the priority of the task is made lower than the priority of at least one other task, by adjusting the priority of each task based on its execution time, excessive execution of the high priority task is prevented, A task priority management system that can improve processing uniformity can be provided.

According to the second aspect of the present invention, the job information storage means storing the job identifier for each job and each task belonging to the same job for each job identifier stored in the job information storage means The task information storage means for storing the identifiers, the respective task priorities corresponding to the respective task identifiers, and the other task identifiers belonging to the same job are provided, and the in-job priority adjustment means stores the task information storage means. When a certain task priority of the task priorities is changed, the priority of another task belonging to the same job as the task having the changed task priority is based on the stored contents of the task information storage means and the job information storage means. The task priority management method that can ensure job consistency by changing the priority so that the order of priorities in the same job is not changed It can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a computer system applied to a task priority management system according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of a job table and a task table in the same embodiment.

FIG. 3 is a flowchart for explaining the operation in the same embodiment.

FIG. 4 is a flowchart for explaining the operation in the embodiment.

FIG. 5 is a flowchart for explaining the operation in the embodiment.

[Explanation of symbols]

1 ... internal timer, 2 ... scheduler, 3 ... OS, 4 ... storage unit, 5 ... CPU, 6 ... job table, 7 ... task information, 8 ... task table.

Claims (2)

[Claims] 1. A computer system in which a storage unit having a plurality of tasks each having a priority is connected to an operation system having an internal timer, and each task in the storage unit executes a process based on the priority. While each task is generated based on the priority, a scheduler that monitors the execution time for each execution of each task is compared with the execution time monitored by this scheduler and a predetermined reference execution time, and the A task priority management method, comprising: priority adjusting means for lowering the priority of the task when compared with the reference execution time when the task execution time exceeds the reference execution time. . 2. An operation system having an internal timer is connected to a storage unit having a plurality of jobs each having a priority, and each task in the storage unit executes a process based on the priority. In a computer system, a job information storage unit that stores a job identifier for each job, each job identifier stored in this job information storage unit, each task identifier that belongs to the same job, and each task identifier When each task priority and the other task identifiers belonging to the same job are stored, and a task priority among the task priorities stored in this task information storage means is changed, The changed task priority is based on the stored contents of the task information storage means and the job information storage means. A task priority management method, comprising: an in-job priority adjusting means for changing the priority of another task belonging to the same job as the own task so as not to change the order of priorities in the same job.