JPH1165859A - Method and device for managing operation program and storage medium where program for operation program management is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent such trouble that the execution of an operation program of high priority is impeded and to greatly improve the system performance by setting operation priority by operation programs to be managed and assigning a CPU to an operation program of high priority more than an operation program of low priority. SOLUTION: A computer device which executes a plurality of operation programs 22 by sharing a CPU 1 set the operation priority by the operation programs 22 in advance and also sets the target use rate of the CPU 1. The load state of the CPU is checked at predetermined check intervals, stop times in periods at the check intervals are calculated by the operation programs according to the target use rate of the CPU 1 and the operation priority of the respective operation programs, and the operation stop and restart of each operation program are controlled according to the stop times. Consequently, the CPU 1 is assigned to the operation programs of high priority more than those of low priority.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for automatically managing a plurality of different business programs that are simultaneously executed by sharing a CPU (Central Processing Unit) in a computer device, and to record a business program management program. Related to a recorded medium.

[0002]

2. Description of the Related Art Conventionally, in a multi-programming type computer device capable of simultaneously executing a plurality of different business programs by sharing a CPU, it has been possible to temporarily stop a running program. But CPU
The management of the start and stop of each business program has not been performed in consideration of the load status of each program and the priority of each business program. For this reason, when the CPU is operating under a high load state, even if a start request for an urgent business program that requires urgency is made, the request may not be accepted.

[0003]

As described above, in the conventional multi-programming type computer apparatus,
Since the start and stop of each business program were not managed in consideration of the load state of the CPU and the priority of each business program, the execution of the low-priority business program hindered the execution of the high-priority business program. However, there is a risk that the system performance will be affected. Therefore, an object of the present invention is to improve system performance by increasing the allocation of CPUs to business programs with high priority.

[0004]

According to the program management method of the present invention, the priority of operation is set in advance for each business program and the target usage rate of the CPU is set. The CPU load status is checked at predetermined check intervals, and each time the stop time within the check interval period is determined for each business program based on the target CPU usage rate and the operation priority of each business program. Is calculated, and the operation stop and restart of each business program are controlled in accordance with the stop time.

[0005] The program management device of the present invention is provided with a program information storage means for storing information relating to the priority of operation set for each business program, and a target usage rate storage means for storing a target usage rate of the CPU. Also, CPU
And a load state detecting means for detecting the load state. The load state of the CPU detected by the detection means is checked at predetermined check intervals, and each time the CPU utilization stored by the target utilization storage means and each task stored by the program information storage means are checked. The stop time within the check interval period is calculated for each business program based on the information on the operation priority of the program, and the operation time of each business program within the check interval period is calculated according to the stop time for each business program. This device has a function to control stop and restart.

A computer-readable recording medium according to the present invention has a function of checking the load state of a CPU at predetermined check intervals in a computer device capable of simultaneously executing a plurality of different business programs while sharing a CPU; C
Each time the load status of the PU is checked, a preset CP
A function of calculating the stop time within the check interval period for each business program based on the target usage rate of U and the operation priority of each business program, and a function of calculating the stop time of each business program according to the stop time calculated for each business program. It records a business program management program for realizing a function of controlling operation stop and restart.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram functionally showing a main configuration of a computer device according to the embodiment. As shown, such a computer device comprises:
A CPU 1 is mounted as a control unit main body, and the CPU 1
Program storage unit 2, RAM (RandomAccess Memoly)
3, an auxiliary storage unit 4, a clock unit 5, and an interface unit 6 are connected by a bus line 7.

The program storage unit 2 stores an operating system (OS) for controlling basic functions of the computer device.
m) 21 and a plurality of different business programs 22 for controlling various businesses executable by the computer device. Here, the OS 21 implements a multi-programming method capable of simultaneously executing a plurality of different business programs 22 while sharing the CPU 1. Further, it has a function of acquiring the current usage rate (%) of the CPU 1.

The RAM 3 is an area for storing various data in a rewritable manner, and in particular, a program management memory 30 described later is created. The auxiliary storage unit 4 stores, for example, an FD (Flop
py Disk) and other magnetic disks and CDs (Compact Discs)
An area for storing information read from a recording medium such as an optical disk such as a ROM and a DVD (Digital Video Disc);
1 and the management information file 42 are read from the recording medium and stored.

The clock section 5 measures the system time, and the CPU 1 has a timer 10 that counts up each time the system time measures 1 second. The interface unit 6 exchanges data with input devices such as a keyboard and a mouse, and output devices such as a display and a printer.

[0011] The computer device having such a configuration is provided with an OS 2
Under the control of 1, the business program management program 41 and the management information file 42 recorded on the recording medium are read and stored in the auxiliary storage unit 4, and the business program management program 41 is started automatically or by operation. Then, the CPU 1 starts the processing shown in the flowchart of FIG.

That is, the CPU 1 first reads the management information file 42 stored in the auxiliary storage unit 4 as ST1 and temporarily stores it in the RAM 3. This management information file 42
As shown in FIG. 3, an interval (second) for checking the load state of the CPU 1, a target usage rate (%) of the CPU 1, and an assignment to each of the business programs 22 to be managed stored in the program storage unit 2. This is a data file in which each piece of information on the assigned operation priority (%) is set in advance.

As the usage rate of the CPU 1 approaches 100%, the system efficiency increases. However, in a state where the usage rate is close to 100%, there is no room to execute the processing immediately for a sudden processing request. Further, it is not known how much the CPU 1 is insufficient if the usage rate is always 100%. The target usage rate of the CPU 1 may be set in consideration of this point.

Each business program 2 to be managed
No. 2 is divided into those that execute urgent and high-immediate processing and those that execute urgent and low-immediate processing. The level of immediacy also varies from one individual to another. Therefore,
The higher the immediateness of the business program, the higher the CPU priority
The priority of each business program is determined so that 1 can be used. Incidentally, the operation priority is a ratio of the CPU use allocation for each business program when the CPU use allocation to the reference business program is set to 100 (%). 50% higher than business programs with 100% operation priority
(1.5 times), which means that the use of the CPU 1 is allocated. There are n (n ≧ 2) business programs 22 to be managed, each of which is managed by assigning a program number of 1, 2,..., N.

Next, the CPU 1 proceeds to step ST2 where the RAM 3
Then, the program management memory 30 is created, and the program management memory 30 is initialized. Here, as shown in FIG. 4, the program management memory 30 includes a storage area 31 for the current usage rate (%) of the CPU 1, a storage area 32 for the target usage rate (%) of the CPU 1, ) And a storage area 34 for the current stop reference time (second).
Storage area 35 for the check interval (second), storage area 36 for the stop time (second) of each business program 22 to be managed, and a flag (0 = stop) indicating the operation status of each business program 22 Medium, 1 = Working, 2 =
(Suspended) storage area 37. When the initialization is performed, all the storage areas 31 to 37 are cleared to [0].

Next, the CPU 1 checks whether or not a request to start the business program 22 has been issued automatically or by operation as ST3. If the activation request has been issued (YES in ST4), ST5
The business program 22 that has been requested to start is started. In step ST6, among the operation status flags for each application program stored in the storage area 37 of the program management memory 30, the operation status flag corresponding to the activated application program 22 is set to "1" (operating). After a while, ST
Returning to step 3, it is checked whether or not an activation request for another business program 22 has occurred. When the started business program ends, the operation status flag corresponding to the business program 22 is returned to “0” (stopped).

The business program 2 is checked by checking ST3.
2 has not been issued (ST4 N
O) Then, as ST7, the current usage rate (%) corresponding to the current load state of the CPU 1 is obtained from the OS 21 and written into the storage area 31 of the program management memory 30. Further, the target usage rate of the CPU 1 read from the operation information file 9 is written in the storage area 32 of the program management memory 30. However, when the usage rate of the CPU 1 is 100% immediately after the start of the business program management program 41, the target usage rate is set to 50%. This makes it possible to quickly set the usage rate of the CPU 1 to an appropriate value.

The CPU 1 then proceeds to step ST8, where the CPU current usage rate (%), the CPU target usage rate (%) stored in the storage areas 31, 32, 33, and 35 of the program management memory 30, respectively, (Seconds) and check interval (seconds). Then, the reference stop time (second) in the current check period is calculated by the following equation [1].
Is calculated. Current reference stop time = Previous reference stop time + (CPU current usage rate-CPU target usage rate) x (check interval-Previous reference stop time) ... [1] (However, if the current reference stop time <0, the current reference stop time After calculating the current reference stop time (seconds), the CPU 1 stores the contents of the storage area 34 (the current reference stop time) in the program management memory 30 in the storage area 33.
And updates the previous reference stop time, and then writes the calculated current reference stop time into the storage area 34.

Next, as ST9, each data of the current reference stop time (second) and the check interval (second) stored in the storage areas 34 and 35 of the program management memory 30 are read. Then, using the operation priority (%) of each business program read from the operation information file 42, the stop time (second) of each business program 22 in the current check period is calculated for each business program by the following equation (2). calculate. Program stop time = check interval− (check interval−current reference stop time) × operation priority (2) (however, if program stop time <0, program stop time = 0)
Is calculated for each business program, the program stop time is written into the storage area 36 of the program management memory 30 for each program.

Next, at ST10, the system time measured by the clock unit 5 is read as a check start time, stored in the RAM 3, and the timer 10 is once reset and restarted. Thereafter, the count value of the timer 10 is stored in the storage area 35 of the program management memory 30.
Until the value of the check interval stored in the service program is reached, that is, until the time corresponding to the check interval elapses from the check start time, each of the business programs according to the stop time calculated for each business program 22 in ST11. Controls operation stop and restart. Then, if this operation stop / restart control process is executed for all the business programs 22 to be managed, and it is confirmed that the time corresponding to the check interval has elapsed from the check start time (ST1).
2 YES), returning to ST3, again in business program 2
Check the start request of 2. If there is a start request, the processing of ST5 and ST6 is executed.
The processes of 7, 8, 9, 10, 11, and 12 are executed again.

FIG. 5 is a flowchart specifically showing the operation stop / restart control processing of each business program 22 in ST11 in FIG. 2. The CPU 1 controls the business program 22 to be managed.
This process is repeated by the number of times. Items A to E relating to various times used in this processing are shown in FIG.
Shows the timing.

First, the operation stop / restart control processing is as follows.
The check interval is equally divided by the number of divisions = X (X = 4 in FIG. 6), one stop period is created in one section for each business program 22, and stop and restart are controlled accordingly. .
In this case, if the suspension periods E of all the business programs 22 to be managed are concentrated from the beginning of one section, all the business programs 22 are stopped in the first half of the section, and the second half of the section is the task for which the start request was issued. The phenomenon that all the programs are restarted occurs. Therefore, in the present embodiment, one division is further divided by the number n of business programs to be managed, and the stop period E of each business program 22 is shifted little by little, so that the CPU 1
We are trying to separate out the use of.

That is, the CPU 1 first resets a counter N built in the CPU to [0] as ST1. Next, if the counter N is incremented by +1 in ST2, it is determined in ST3 whether the counter N has exceeded the number n of the business programs 22 to be managed. If not exceeded, the system time is read from the clock unit 5 as ST4, and the elapsed time A from the check start time acquired in the process of ST10 in FIG. 2 is calculated.

Next, at ST5, the check interval is read from the storage area 35 of the program management memory 30, and the quotient obtained by dividing the check interval by the number of divisions X is obtained. And
The elapsed time A is divided by this quotient, and the remainder is obtained as the elapsed time B in one section. In ST6, the quotient obtained by dividing the check interval by the number of divisions X is further divided by the number n of the business programs 22 to be managed, and the quotient is calculated as the program shift time C in one section. further,
In ST7, the program shift time is multiplied by a value obtained by subtracting 1 from the counter N, and the product is calculated as the stop start time D in one section of the business program whose program number matches the counter N.

Thereafter, as ST8, the stop time E of the business program whose program number matches the counter N is read from the storage area 36 of the program management memory 30, and it is determined whether the following evaluation formula 1 is satisfied. Evaluation formula 1: (program stop time E / division number X) <(check interval / division number X) -stop start time D That is, the time E that the business program should stop during one division is the stop start during the division. It is determined whether the time is smaller than the time from the time D to the end of the division. If it is smaller, the process proceeds to ST9, and if it is larger, ST9
Go to 10.

In both ST9 and ST10, it is determined whether or not the following evaluation formula 2 is satisfied. Evaluation Formula 2: Elapsed time B in one section ≧ stop start time D That is, it is determined whether or not the current time in one section has passed the stop start time in that section. And S
If it is determined at T9 that the time has passed, the process proceeds to ST12, and if it is determined that the time has not passed, the process proceeds to ST13.
If it is determined in ST10 that the time has passed, ST
Proceed to step 12 and if it is determined that the time has not passed, ST11
Proceed to.

In ST11, the stop time E of the business program whose program number matches the counter N is read from the storage area 36 of the program management memory 30 again, and it is determined whether or not the following evaluation formula 3 is satisfied. Evaluation formula 3: Elapsed time B in one section <(check interval ÷ division number X) − (program stop time E ÷ division number X) That is, elapsed time B in one section is calculated from stop start time D in that section. It is determined whether or not the time is shorter than the time until the end of the classification. If it is determined to be smaller, the process proceeds to ST12, and if it is determined to be larger, the process proceeds to ST13.

In ST12, the operation status flag of the business program whose program number matches the counter N is read from the storage area 37 of the program management memory 30. When the operation status flag is "1" (operating), ST
At 14, the operation of the business program is stopped. In ST15, among the operation status flags for each business program stored in the storage area 37 of the program management memory 30, the operation status flag corresponding to the business program 22 whose operation has been stopped is set to “2” (interrupted). If the operation status flag of the relevant business program is “0” (stopped) or “2” (interrupted) in ST12, ST14, ST1
Step 5 is not performed. Thereafter, returning to ST2, the counter N is counted up by +1. And the counter N
If has not yet exceeded the number n of business programs, the processing after ST4 is repeated.

On the other hand, at ST13, the operation status flag of the business program whose program number matches the counter N is read from the storage area 37 of the program management memory 30. If the operation status flag is "2" (interrupted), the operation of the business program is restarted in ST16. Also, as ST17, the program management memory 30
Among the operation status flags for each business program stored in the storage area 37, the operation status flag corresponding to the business program 22 whose operation has been resumed is returned to “1” (in operation). S
If the operation status flag of the relevant business program is “0” (stopped) or “1” (operating) at T13,
The processing of ST16 and ST17 is not performed. Thereafter, returning to ST2, the counter N is counted up by +1. If the counter N has not yet exceeded the number n of business programs, the processing after ST4 is repeated.

In the computer apparatus according to the present embodiment configured as described above, a management information file 42 is separately created based on the business program 22 to be managed, and recorded on a storage medium together with the business program management program 41. . Then, the present invention is realized by reading the recording medium into the computer device. Note that the management information file 42 and the business program management program 41 may be recorded in different storage media, and each may be read by a computer device.

The business program 22 to be managed now
And the operation priority of the business program of program number 1 written in the management information file 42 is set to 120.
%, The operation priority of the business program with the program number 2 is 100%, the operation priority of the business program with the program number 3 is 150%, the operation priority of the business program with the program number 4 is 80%, and the business program with the program number 5 Is set to 50%. Further, the target usage rate of the CPU 1 is set to 90%, and the check interval is set to 100 seconds.

Now, the five business programs 1 to be managed
~ 5 have been activated and the usage rate of CPU1 is 100
%, The business program management program 41
Are activated, the contents of the program management memory 30 change as shown in the records 81, 82,..., 88 in FIG. 8 during the first to eighth check periods.

That is, in the first check period, C
The target usage rate of PU1 is set to 50%, and the reference stop time this time is 50 seconds (100 seconds × 50%) according to the formula [1].
Is calculated. In addition, the stop time of each business program 22 is calculated as follows by the above equation [2]. Stop time of the business program of program number 1 = 100
Second-(100 seconds-50 seconds) x 120% = 40.0 seconds Stop time of the business program of program number 2 = 100
Second-(100 seconds-50 seconds) x 100% = 50.0 seconds Stop time of the business program of program number 3 = 100
Second− (100 seconds−50 seconds) × 150% = 25.0 seconds Stop time of the business program of program number 4 = 100
Second-(100 seconds-50 seconds) x 80% = 60.0 seconds Stop time of the business program of program number 5 = 100
Sec-(100 sec-50 sec) x 50% = 75.0 sec.

In the first check period, the business program of the program number 1 having the operation priority of 120% is stopped for 40 seconds during the check period of 100 seconds, and the business program of the program number 2 having the operation priority of 100% is performed. The program is stopped for 50 seconds during the check period of 100 seconds, and the business program of the program number 3 with the operation priority of 150% is stopped for 25 seconds during the check period of 100 seconds, and the business program of the program number 4 with the operation priority of 80% is Check period 10
A business program with a program number 5 that stops for 60 seconds in 0 seconds and has an operation priority of 50% is 75 in 100 seconds in the check period.
Pause for seconds.

By the way, as shown in FIG. 6, the timing at which the operation of each business program stops is determined by the check interval 10
For example, every 25 seconds obtained by dividing 0 seconds into four, FIG.
As shown in (1), stop and restart are controlled by shifting by the shift time C for each business program. In FIG. 7, E1 indicates the suspension period of the business program of program number 1, E2 indicates the suspension period of the business program of program number 2, and E3 indicates the suspension period of the business program of program number 3. E4 indicates the suspension period of the business program of program number 4, and E5 indicates the suspension period of the business program of program number 5. Thus, the suspension period E of each business program 22
By shifting the E1, E2, E3, E4, and E5 little by little, the use of the CPU 1 can be determined in a somewhat balanced manner.

Here, the usage rate of the CPU 1 is still 1
If it is 00%, even in the second check period, C
The target usage rate of PU1 is set to 50%. Thus, the current reference stop time is calculated as 75 seconds. The stop time of each business program 22 is 70.0 seconds, 75.0 seconds, 62.5 seconds, 80.0 seconds in the order of the program numbers.
Seconds, 87.5 seconds.

Thus, in the second check period, the business program of the program number 1 stops for 70 seconds in the 100-second check period, the business program of the program number 2 stops for 75 seconds in the 100-second check period, The business program of program No. 3 stops for 62.5 seconds during the check period of 100 seconds, the business program of program number 4 stops for 80 seconds of the check period of 100 seconds, and the business program of program number 5 stops for 87.
Pause for 5 seconds.

That is, the stop time of each business program 22 is longer than in the first check period. As a result, if the usage rate of the CPU 1 is reduced to 70%,
In the third check period, the target usage rate of the CPU 1 is set to 90% in the management information file 42. As a result, the current standard stop time is calculated as 70 seconds. The stop time of each business program 22 is 64.0 seconds, 70.0 seconds, 55.0 seconds, 76.0 seconds in the order of the program numbers.
Second, 85.0 seconds.

Thus, in the third check period, the business program of the program number 1 stops for 64 seconds in the 100-second check period, the business program of the program number 2 stops for 70 seconds in the 100-second check period, The business program No. 3 stops for 55 seconds during the check period 100 seconds, the business program No. 4 stops for 76 seconds during the check period 100 seconds, and the program number 5
Business program stops for 85 seconds during the check period of 100 seconds.

That is, the stop time of each business program 22 is shorter than in the second check period. As a result, if the usage rate of CPU1 rises to 95%,
In the fourth check period, the current standard stop time is 71.
It is calculated as 5 seconds. Also, the stop time of each business program 22 is 65.8 seconds in the order of the program numbers, and 71.
It is calculated as 5 seconds, 57.3 seconds, 77.2 seconds, and 85.8 seconds.

Thus, in the fourth check period, the business program of the program number 1 is stopped for 65.8 seconds during the check period of 100 seconds, and the business program of the program number 2 is stopped for 71.5 seconds during the check period of 100 seconds. The business program of program number 3 is stopped for 57.3 seconds during the check period of 100 seconds, the business program of program number 4 is stopped for 77.2 seconds during the check period of 100 seconds, and the business program of program number 5 is stopped for 100 check periods. 85.8 seconds pause in seconds.

That is, the stop time of each business program 22 is slightly longer than that in the third check period. As a result, if the usage rate of the CPU 1 matches 90% of the target value, the same result as the fourth check period is obtained in the fifth check period. Thereafter, in the same manner, the sixth, seventh, and eighth checks are periodically performed. Each time, the stop time within the check period is calculated for each business program 22, and the running program is stopped and restarted. Control is performed.

As described above, according to the present embodiment, the CP
A plurality of different business programs 22 sharing U1 at the same time
In the computer device capable of executing the operations, the operation priority is set in advance for each business program 22 and the CPU 1
Set a target usage rate for Then, the CPU 1 checks the load state of the CPU 1 at predetermined check intervals.
A stop time within a check interval period is calculated for each business program based on the target usage rate of the PU1 and the operation priority of each business program, and operation stop and restart of each business program are controlled according to the stop time. I have to.

As a result, the assignment of the CPU 1 to the high-priority business program is larger than that of the low-priority business program. Therefore, there is no problem that the execution of the low-priority business program prevents the execution of the high-priority business program, and the system performance can be improved. Further, there is an effect that the load on the CPU 1 can be prevented from increasing due to a bug or an abnormal operation of the business program.

The business program management program 41 described above is conveniently configured to always monitor the load state of the CPU 1 and to automatically execute the program only when the load is high.

[0046]

As described in detail above, according to the present invention,
Only by setting the operation priority of the business program to be managed for each program, the assignment of the CPU 1 to the high-priority business program can be made larger than that of the low-priority business program. It is possible to prevent a problem that the execution of a program with a high priority is hindered by the execution of the program, and to achieve an excellent effect that the system performance can be greatly improved.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing a main configuration of a computer device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a processing procedure of a business program management program according to the embodiment.

FIG. 3 is an exemplary view showing a data structure of a management information file according to the embodiment;

FIG. 4 is an exemplary view showing an area structure of a program management memory according to the embodiment;

FIG. 5 is a flowchart specifically showing stop / restart control processing for each business program in FIG. 2;

FIG. 6 is a diagram used to explain items related to time used in the processing procedure in FIG. 5;

FIG. 7 is a diagram showing an example of the timing of stop / resume control of each business program.

FIG. 8 is a diagram showing an example of transition of records stored in a program management memory.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... CPU 2 ... Program storage part 21 ... OS 22 ... Business program 3 ... RAM 30 ... Program management memory 4 ... Auxiliary storage part 41 ... Business program management program 42 ... Management information file 5 ... Clock part

Claims (4)

[Claims] 1. A business program management method for a computer device capable of simultaneously executing a plurality of different business programs by sharing a CPU, wherein a priority of operation is set in advance for each of the business programs and a target use of the CPU is controlled. A load rate of the CPU is checked at predetermined check intervals, and a period of the check interval is set for each business program based on the target usage rate of the CPU and the operation priority of each business program each time. A method of managing business programs, comprising calculating a stop time in a computer, and controlling operation stop and restart of each business program according to the stop time. 2. The business program management method according to claim 1, wherein when the operation of each business program is stopped according to the calculated fixed stop time, the operation stop timing is shifted for each business program within a check interval. A business program management method, characterized in that: 3. A business program management device which manages a plurality of different business programs executed simultaneously by sharing a CPU, wherein: a program information storage means for storing information relating to an operation priority set for each business program; A target usage rate storage means for storing a target usage rate of the CPU; a load status detection means for detecting a load status of the CPU; a predetermined check interval for the load status of the CPU detected by the detection means. The check interval is set for each business program based on the target usage rate of the CPU stored by the target usage rate storage means and the information on the operation priority of each business program stored by the program information storage means each time. Calculation means for calculating the stop time within the period, and work calculated by the calculation means Wherein within a period of the check interval in response to the stop time of each program operation stop of the service program, business program management apparatus characterized by comprising a program operation control means for controlling restart. 4. A function of checking the load status of the CPU at predetermined check intervals in a computer device that can execute a plurality of different business programs simultaneously while sharing a CPU, and checks the load status of the CPU. A function of calculating a stop time within the period of the check interval for each business program based on a target usage rate of the CPU and an operation priority of each business program preset for each business program; A computer-readable recording medium for recording a business program management program for realizing the function of controlling the operation stop and restart of each business program according to the stop time.