JPH09160792A - Data processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten the processing time of an optional task so as to finish it within a desired time while plural tasks are processed in time division and in parallel to each other. SOLUTION: A multitask management table 15-1 stores the assigned use ratio to show the processing rate of every task per unit time among those tasks which are processed in time division and in parallel to each other. A CPU 11 refers to the table 15-1 to process the tasks in time division and in parallel to each other according to every assigned use ratio of the tasks. When the input is designated for the processing request time to request the processing end time point of the corresponding task at its starting time, the CPU 11 increases the assigned use ratio of every task in order to complete the processing of the relevant task.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing device which processes a plurality of tasks in parallel in a time division manner.

[0002]

2. Description of the Related Art Conventionally, in a data processing device having a scheduler function for starting an automatic program, when a start time of a business process is reached, the business process is executed. In this case, when another task process is started by the scheduler function during execution of a task, a plurality of tasks are processed in parallel in a time-sharing manner. When a plurality of tasks are time-divisionally processed in parallel in this way, the end point of each task greatly varies depending on the processing content of the task and the number of tasks.

[0003]

By the way, when a plurality of tasks are processed in parallel in a time-sharing manner, the weight (urgent level) of each task is not always uniform, and some tasks require urgent processing depending on the business content. There is also. For example, the next business process (for example, the form printing process) cannot be started until the update of the data file is completed, and therefore the speed of the file update process is required. In such a case, other processing during parallel processing is interrupted, but it is not always the best method depending on the business content. Especially, data transmission / reception between the host computer and multiple terminal devices In the data communication system that performs the above, the host computer performs various kinds of processing in parallel in a time-divisional manner in response to the request of each terminal, but if a certain processing is interrupted, the business on the terminal side stops, May cause major operational problems. An object of the present invention is to be able to shorten the processing time so that a plurality of tasks can be completed within a desired time with respect to an arbitrary task while being concurrently processed in a time-sharing manner.

[0004]

The means of the present invention are as follows. In a data processing device that processes a plurality of tasks in parallel in a time-sharing manner, (1) the usage rate storage means allocates to each task that is concurrently processed in a time-sharing manner, at what rate it is being processed per unit time. Memorize the usage rate. (2) The processing execution means processes a plurality of tasks in parallel in a time-division manner according to the assigned usage rate of each task stored in the usage rate storage means. (3) The input means inputs the end timing information for requesting the end point of the processing in association with any task. The end timing information is the end time and the time until the end of processing. (4) Based on the end timing information input from the input means and the processing progress state up to now, the determining means determines whether or not the task ends by the time point indicated by the end timing information. (5) When the determination means determines that the task is not finished, the task control means increases the current allocation usage rate for the task and changes the allocation usage rates of the other tasks accordingly. The input means may be a communication control means for receiving the end timing information transmitted from the terminal device. Now, for example, when the end timing information for requesting the end point of the processing is input at the time of starting the task, the determining means determines the end timing information input from the input means and the processing progress state up to the present time. It is determined whether the task ends by the time point indicated by the end timing information. As a result, when it is determined that the task is not finished, the task control unit increases the current allocation usage rate for the task and changes the allocation usage rates of other tasks. Therefore, the processing time can be shortened so that a plurality of tasks can be completed within a desired time with respect to an arbitrary task while being concurrently processed in a time-sharing manner.

[0005]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. FIG. 1 shows a host computer 1 on the server side and a terminal device 2 on the client side.
3 is a block diagram showing a configuration of a host computer 1 in a data communication system for transmitting and receiving data to and from. The CPU 11 is a central processing unit that controls the overall operation of the host computer 1 in accordance with various programs stored in the ROM 12 and the like, and stores data fetched from the disk 14 via the disk control unit 13 in the RAM 15 or The data in the RAM 15 is registered and stored in the disk 14, and the data is transmitted / received to / from the terminal device 2 via the communication control unit 16.

The CPU 11 is capable of processing a plurality of tasks in parallel in a time-sharing manner, and the ROM 1 is operated in accordance with a processing start command input from the key input unit 17 or the communication control unit 16.
When the processing program in 2 is started or the processing start time is detected by the scheduler, the processing program in the ROM 12 is automatically started. Here, when a plurality of tasks are processed in parallel in a time-sharing manner, the CPU 11 is
The multitask is executed by referring to the multitask management table 15-1 and the multitask parameter memory 15-2 in FIG.

The multitask management table 15-1 is shown in FIG.
As shown in FIG. 5, the boot program name and the allocation usage rate are stored in association with each other. Here, the assigned usage rate indicates how much of each task is processed per unit time, and the value is written at predetermined time intervals (for example, every 1 minute) as needed during multitask execution. Can be replaced. In other words, RAM1 for each task
The allocation usage rate set to 5 is normally set to a uniform value according to the number of tasks, but when the allocation usage rate of a task is changed, the allocation usage rate of other tasks is correspondingly changed. Is also changed, and the change is performed every predetermined time. This multitask management table 15
The CPU 11 processes a plurality of tasks in parallel in a time-sharing manner according to the allocation usage rate of each task stored in -1.
That is, the CPU 11 determines at what rate each task is processed per unit time by the multi-task management table 15.
It decides based on the corresponding allocation utilization in -1, and according to this, several tasks are processed in parallel in time division.

The multitask parameter memory 15-2 stores various parameters as shown in FIG.
This parameter is variable data for obtaining the allocation usage rate for each task, and the CPU 11 refers to the multitask parameter memory 15-2 to rewrite the contents (allocation usage rate) of the multitask management table 15-1. Here, the parameters REC1 to REC4 are the number of records (the number of records), and the parameters TIME1 to
TIME3 is a time parameter, the content of which is defined as shown in FIG.

Next, the operation of the host computer 1 will be described with reference to FIG.
And it demonstrates according to the flowchart shown in FIG. FIG. 4 is a flow chart showing the operation in the case of managing multi-tasks, and the execution is started when the recording of a certain business process is designated. First, when the writing of the business process is designated, the processing program is started and the multitask management program is also started. According to this multitask management program, the CPU 11
First, the name of the startup program is the multitask management table 1
5-1 and the allocation utilization rate are set in the multitask management table 15-1 (step A1).
In this case, the allocation usage rate is usually a value evenly divided according to the current number of multitasks. Now, if the number of tasks is "2", "50%" is set as the allocation utilization rate, but even if the number of tasks is "2", a value less than 50% may be set. In other words, the value will be a value according to the allocation usage rate of the other task, but will normally be a value evenly divided.

Next, at the time of starting a business, a time for requesting the end of this processing (in the case of an urgent business, a time required for processing the entire designated business is requested. Time) is input from the key input unit 17,
It is checked whether the data has been transmitted from the terminal device 2 (step A2). Here, if the processing request time of the designated work is not input and designated, the process proceeds to step A18, and the starting program name and the assigned usage rate of the processing are deleted from the multitask management table 15-1. When the processing request time of the designated work is input and designated, the process proceeds to step A3, the contents of the multitask parameter memory 15-2 are initialized, and each parameter is set to "0".

Then, the CPU 11 accesses the processing target file of the designated job, detects the number of processing (for example, the number of records) of the entire designated job, and sets this number of processing as a parameter R.
EC1 is set in the multitask parameter memory 15-2 (step A4), and the input request processing time (in minutes) is fetched, and the parameter TIM is set.
It is set in the multitask parameter memory 15-2 as E1 (step A5). Now, it is assumed that "66000" is set as the total number of processings of the specified work and 3 hours (180 minutes) is set as the total processing request time of the specified work in the multitask parameter memory 15-2.

Then, the CPU 11 has a fixed time (for example, 1
For a period of time), the multitask management process is waited (step A6). During this time, business processing progresses, and when this causes the number of processed cases to be transferred from the business processing routine,
This processed number is stored in the multitask parameter memory 15
-2 (step A7). That is, the parameter REC2 is the cumulative number of processed records up to the present time,
A process of adding the number of processed items this time to the value of the parameter REC2 is performed. Next, in order to obtain the number of processings per predetermined time (1 minute) this time, the value of the parameter REC3 is subtracted from the value of the parameter REC2, and the calculation result is set in the multitask parameter memory 15-2 as the parameter REC3. (Step A8). here,
Initially, the parameters REC2 and REC3 are each set to "0" by the initialization process of step A3, so the number of processed items in the first minute is the parameter REC2.
Is set to (step A7), the parameter REC
The value of 3 becomes the value of the parameter REC2 as it is.

Next, the CPU 11 subtracts the value of the parameter REC2 from the value of the parameter REC1 in order to obtain the number of unprocessed items up to the present time, and the calculation result is used as the parameter REC4 in the multitask parameter memory 15-2.
(Step A9). Now, assuming that the number of processing in the first one minute is "300", "65700" is required as the remaining number. Further, a process of adding a predetermined time (1 minute) to the value of the parameter TIME2 is performed in order to obtain the cumulative processing time up to the present time (step A10). As a result, the parameter TIME2 becomes the cumulative time (in minutes) required for the processing up to the present time,
From the value of the parameter TIME1, this parameter TIME
The remaining processing time up to the present time is obtained by subtracting the value of 2 and set in the multitask parameter memory 15-2 as the parameter TIME3 (step A1).
1). Therefore, when the first one minute has elapsed, the value of the parameter TIME3 becomes "179" minutes.

When the processing for calculating various parameters is completed in this way, the process proceeds to step A12, where it is checked whether a business end notification has been transferred from the business processing routine. Now, when the first one minute has elapsed. Therefore, the process proceeds to step A13, and in order to complete the designated job processing within the remaining time, the number of cases to be processed per minute in the future is calculated. That is, the parameter REC4 is the number of unprocessed items up to the current time, and the parameter TIME3 is the remaining processing time up to the current time.
By dividing the value of C4 by the value of the parameter TIME3, the number of processings per minute to be processed within the remaining time is obtained, and is set in the multitask parameter memory 15-2 as the parameter REC5. Currently, the number of unprocessed items up to the present time is “65700”, and the remaining processing time is “1”.
Since it is 79 "minutes," 367 cases / minute "is obtained as the number of processing cases per minute to be processed within the remaining time.

Then, the value of the parameter REC3 and the value of the parameter REC5 are compared (step A14). As a result, if the condition of "REC3 <REC5" is satisfied, the process proceeds to steps A15 to A17, but if not satisfied. If A
15 to A17 are skipped. That is, the parameter R
EC3 is the number of processings per minute this time, parameter REC5 is the number of processings per minute to be processed within the remaining time, and step A14 is the parameter REC3,
By comparing the values of REC4, it is checked whether or not the number of processings this time is sufficient in order to complete the processing within the remaining time. Here, if “REC3 <REC5”, the number of processings this time is not enough and it is necessary to increase the number of processings per minute in the future.
Move on to execution of A17. First, the CPU 11 sets the parameter R
Calculate the ratio between EC3 and REC5 (Step A
15). Now, the number of processing per minute this time is "30
Since the number of cases is “0” and the number of cases to be processed within the remaining time per minute is “367”, the increase rate of the number of cases is “22%”. Next, the CPU 11 accesses the multitask management table 15-1 and fetches the allocation usage rate (50% in this case) of the current task (step A16). Then, the process of changing the allocation usage rate is performed (step A17).

FIG. 5 is a flowchart showing the allocation utilization rate changing process. First, in order to obtain a new allocation usage rate, the CPU 11 adds the share x calculated in step A15 to the allocation usage rate of the current task acquired in step A16, and sets the result as a new allocation usage rate. (Step B1). Now, a value "72%" obtained by adding "22%" as a share x to the current allocation utilization rate "50%" is obtained as a new utilization rate y. If the new usage rate y does not exceed 100% (step B2), the contents of the multitask management table 15-1 are rewritten based on the usage rate y (step B).
3). That is, the usage rate of the current task is rewritten from "50%" to "72%", and the usage rates of other tasks are also rewritten from "50%" to "100-72 = 28%". If the value of y exceeds 100% in step B2, the change processing of the multitask management table 15-1 is not performed.

When the modification process of the multitask management table 15-1 is completed in this way, the process returns to step A6 in FIG. 4 and the above operation is repeated every minute thereafter. Then, when the business processing end notification is received from the business processing routine (step A12), the CPU 11 erases the program name and the usage rate from the multitask management table 15-1 (step A18). In this case, if there is another task that is currently activated, the usage rate changing process is also performed. Since the processing of one task is completed, the usage rate of the remaining tasks usually has a value according to the current number of tasks.

As described above, the processing time can be shortened so that a plurality of tasks can be completed within a desired time with respect to an arbitrary task while being concurrently processed in a time-sharing manner. As for the required processing, it is possible to prioritize other tasks and increase the processing speed. Since the multitask management table 15-1 is provided, it is possible to easily monitor the multitask.

If the processing is not completed within the designated processing request time, the fact may be notified to the terminal device 2. In this case, the predicted end time may be obtained and notified to the terminal device 2.

[0020]

As described above, according to the present invention, the processing time can be shortened so that a plurality of tasks can be completed within a desired time for an arbitrary task while being concurrently processed in a time-sharing manner. With regard to urgent processing, it becomes possible to prioritize other tasks and increase the processing speed.

[Brief description of the drawings]

FIG. 1 is a block configuration diagram of a host computer 1.

FIG. 2 is a diagram showing a configuration of a multitask management table 15-1.

FIG. 3 is a diagram for explaining the contents of a multitask parameter memory 15-2.

FIG. 4 is a flowchart showing a multitask management process.

5 is a flowchart showing step A17 (usage rate changing process) of FIG. 4. FIG.

[Explanation of symbols]

 1 Host Computer 2 Terminal Device 11 CPU 12 ROM 15 RAM 15-1 Multitask Management Table 15-2 Multitask Parameter Memory 16 Communication Control Section 17 Key Input Section

Claims (2)

[Claims] 1. A data processing apparatus that processes a plurality of tasks in parallel in a time-sharing manner, and for each task that is concurrently processed in a time-sharing manner, stores the allocation utilization rate at what rate is being processed per unit time. Utilization rate storage means, processing execution means for performing parallel processing of a plurality of tasks in time division according to the assigned utilization rate for each task stored in the utilization rate storage means, and processing associated with any task Input means for inputting the end timing information for requesting the end time point, and based on the end timing information input from the input means and the processing progress state up to the present time, by the time point when the task is indicated by the end timing information A determination unit that determines whether or not to end the task, and a current assignment to the task when the determination unit determines that the task does not end. With increasing utilization, data processing apparatus characterized by comprising a task control means for changing accordingly the allocation usage other tasks. 2. The data processing apparatus according to claim 1, wherein the input means is a communication control means for receiving the end timing information transmitted from the terminal device.