JPH11272511A - Device and method for performance information sampling and recording medium for recording program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To sample performance data which are detailed only in a necessary time band by specifying performance sampling items of a whole system by a job program. SOLUTION: A job starting means 301 has an operation terminal 60 analyze a job control language(JCL) 20 as a job program stored in a JCL storage file 10, a job completion means 303 releases used resources, a performance monitoring item update means 302 generates or updates a performance monitoring item change table 312, a performance information sampling means 304 refers to a performance monitor predefined value table 313 or the performance monitoring item change table 312 and decides an item with regard to which the performance information is sampled, a timer 305 activates the performance information sampling means 304 at every constant interval of time, a monitoring identification flag automatic update means 306 updates the job JCL 20 in the JCL storage file 10 by following an instruction from am operation terminal 50 and performs addition and deletion of a monitoring item identification flag 201.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a performance information collecting apparatus, a performance information collecting method, and a recording medium for recording a program, and more particularly to an identifier for specifying a performance monitoring item in a language describing a procedure of a business program. The present invention relates to a performance information collecting apparatus, a performance information collecting method, and a recording medium for recording a program, wherein a performance information collecting type can be controlled.

[0002]

2. Description of the Related Art Conventionally, performance monitoring of a computer employs a method of collecting information according to preset items. There is also a conventional technique in which information collection is changed by a command during operation of a system by an operation of an operator. Further, “Japanese Unexamined Patent Application Publication No. 06-0679
The general-purpose computer load monitoring method described in "No. 38" is a technique in which a "threshold" is provided for a performance monitoring item (for example, CPU usage rate), and information is collected only when the value exceeds that value.

[0003]

In the above-mentioned prior art, it is difficult for the operator to change the information collection items while being aware of the load being executed.

In the technology described in Japanese Patent Application Laid-Open No. 06-0679938, the system is used effectively, and if the usage rate is always 100%, the "threshold value" is set. There is a problem that it cannot be provided.

[0005] Further, when a business program is executed for a certain long time, and there is a peak in the load usage rate intermittently while the business is running,
Since data is collected everywhere, it is not possible to grasp the entire operation, and there is a problem that even if a performance problem occurs, the information is insufficient.

Further, it is necessary to periodically hold as much performance information as possible, but when the number of collected information items is increased, a large area for storing the information is required. There is a problem.

[0007] An object of the present invention is to dynamically change the items of performance collection information with some restrictions so that a large amount of data can be collected in a situation where information is required even if information is not normally collected. It is to be.

[0008]

A first performance information collecting apparatus according to the present invention comprises: a JCL storage file for storing a job JCL; a computer system; a performance information storage file; a first operation terminal; The job JCL includes a monitoring item identification flag indicating a type of performance information to be collected, and the computer system includes:
A management table including a storage device, a job start unit, a performance monitoring item update unit, a job end unit, a performance information collection unit, a timer, and a monitoring identification flag automatic update unit, wherein the storage device stores control information for each business program A performance monitoring item change table that stores information on the type of performance monitoring item to be monitored that changes with the execution of the business program, and a default performance monitoring default value table that indicates the type of performance information to be always collected. , (A) the job start means transmits the JCL from the first operation terminal.
Analyzing the job JCL as a business program stored in a storage file, (b) the job ending means:
When the execution of the business program ends, the used resources are released, and (c) the performance monitoring item update unit is started from the job start unit and the job end unit, and the performance monitoring item change table is released. (D) the performance information collecting means determines an item to collect performance information with reference to the performance monitoring default value table and the performance monitoring item change table, and (e).
The timer activates the performance information collecting means at regular time intervals, and (f) the monitoring identification flag automatic updating means refers to the contents of the performance information storage file to perform the second processing.
, The job JCL in the JCL storage file is updated, and the monitoring item identification flag is added or deleted.

[0009] A second performance information collecting apparatus according to the present invention is the first performance information collecting apparatus, wherein the job start means analyzes the contents of the business program and executes the job J.
It is determined whether the monitoring item identification flag is described in the CL. If the job JCL has an unanalyzable error, the process is interrupted. If there is no error, the monitoring item identification flag is Is described, the performance monitoring item updating means executes the processing, and then the job starting means copies the contents of the monitoring item identification flag to the management table in the storage device as management information.

A third performance information collecting apparatus according to the present invention is the first or second performance information collecting apparatus, wherein the job ending means refers to the management table for a business program to be ended, and It is determined whether or not the monitoring item identification flag is designated, and when the business program in which the monitoring item identification flag is designated ends, the performance monitoring item updating means executes processing.

A fourth performance information collecting apparatus according to the present invention is the first, second or third performance information collecting apparatus, wherein the performance monitoring item change table has an item name for each monitoring item for which performance information is to be collected. Includes settings and number of requested jobs.

A fifth performance information collecting apparatus according to the present invention is the fourth performance information collecting apparatus, wherein the performance monitoring item updating means is executing processing by the job starting means, or It is determined whether or not the processing by the ending means is being executed. (A) If the processing by the job starting means is being executed, the contents of the monitoring item identification flag described in the job JCL of the business program Is retrieved, the performance monitoring item change table is searched using the content of the retrieved monitoring item identification flag as a key, and when an item name that matches the specified content of the monitoring item identification flag is detected, If the number of requested jobs for the corresponding item name in the monitoring item change table is incremented by "1" and no matching item name is detected, the performance monitoring item change table is added. Create a new entry in, to add a new the item name in the item name, "O the setting
N ", the number of requested jobs is set to" 1 ", and it is determined whether the processed item is the last designated item. If not, the process proceeds to the process of extracting one content of the monitoring item identification flag. Return, if it is the last, end the processing,
(B) When the processing by the job ending means is being executed, the management table of the business program to be ended is searched, and one monitoring item corresponding to the monitoring item identification flag is fetched. The performance monitoring item change table is searched using the item as a key, and “1” is subtracted from the number of requested jobs corresponding to the corresponding monitoring item. If the number of requested jobs becomes “0”, the setting is changed. "O
FF ", and if the processed monitoring item identification flag is the last one, the process is terminated; otherwise, the process returns to the process of extracting one monitoring item corresponding to the monitoring item identification flag. The performance information collecting apparatus according to claim 4, wherein

A sixth performance information collecting apparatus according to the present invention is the sixth performance information collecting apparatus, wherein the performance monitoring default value table corresponds to the content of the monitoring item identification flag for each predetermined value setting item. An item name, which is a name for identifying an item for performance monitoring, and two items of “ON” and “OFF”
And "OFF" if the user does not want to collect the information of the corresponding default value setting item.

A seventh performance information collecting apparatus according to the present invention is the sixth performance information collecting apparatus, wherein the performance information collecting means is activated by being notified by the timer at regular intervals, and Reading one of the default value setting items from a default value table, and determining a value of the corresponding setting;
(A) When the value of the setting in the performance monitoring default value table is “OFF”, an item name equal to the item name corresponding to the read default value setting item is read from the performance monitoring item change table. Read, if the corresponding item name is present in the performance monitoring item change table and the setting in the performance monitoring item change table is “ON”, the performance information corresponding to the item name is collected; (B) When the value of the setting in the performance monitoring default value table is “ON”, performance information corresponding to an item name in the performance monitoring default value table is collected. And saves it in the performance information storage file, and (c)
If the predetermined value setting item extracted from the performance monitoring default value table is not the last, the process returns to the process of reading one of the predetermined value setting items from the performance monitoring default value table, and ends if it is the last.

An eighth performance information collecting apparatus according to the present invention is the seventh performance information collecting apparatus, wherein the monitoring identification flag automatic updating means includes a condition and a change content for updating the JCL from the second operation terminal. And referring to the performance information information of the business program in the performance information storage file,
The job program matching the condition is detected, and a monitoring item identification flag specified by the change content is added to the job JCL corresponding to the job program.

The performance information collecting method according to the present invention uses the job JC
L, a JCL storage file, a computer system, a performance information storage file, a first operation terminal,
A second operation terminal, wherein the job JCL includes a monitoring item identification flag indicating a type of performance information to be collected;
The computer system includes a storage device, a job start unit, a performance monitoring item update unit, a job end unit, a performance information collection unit, a timer, and a monitoring identification flag automatic update unit, and the storage device stores control information for each business program. A performance monitoring item change table that stores information on the type of performance monitoring item to be monitored that changes with the execution of the business program, and a default performance monitor that indicates the type of performance information that should always be collected A performance information collection method for a performance information collection device including a default value table, wherein: (a) the job starting means includes a job as a business program stored in the JCL storage file from the first operation terminal; Analyzing the JCL; and (b) executing the job program by the job ending means. (C) causing the performance monitoring item updating means to be activated by the job starting means and the job ending means to generate the performance monitoring item change table. Updating; (d) causing the performance information collection means to determine an item for which performance information is to be collected with reference to the performance monitoring default value table and the performance monitoring item change table; and (e) the timer. Activating the performance information collecting means at regular time intervals; and (f) instructing the monitoring identification flag automatic updating means to refer to the contents of the performance information storage file and to designate from the second operating terminal Updating the job JCL in the JCL storage file to add or delete the monitoring item identification flag. .

A recording medium according to the present invention includes: a JCL storage file for storing a job JCL; a computer system;
The job JCL includes a performance information storage file, a first operation terminal, and a second operation terminal, wherein the job JCL includes a monitoring item identification flag indicating a type from which performance information is to be collected. A management table that includes an apparatus, a job start unit, a performance monitoring item update unit, a job end unit, a performance information collection unit, a timer, and a monitoring identification flag automatic update unit, wherein the storage device stores control information for each business program; A performance monitoring item change table that stores information on the types of performance monitoring items to be monitored that change with the execution of the business program, and a performance monitoring default value table that indicates the types of performance information that should always be collected (A) For the information collection device, the job start means is stored in the JCL storage file from the first operation terminal. A step of analyzing said job JCL as business program, to (b) the job end unit, when the execution of the service program is completed, a step of causing the release of resources used,
(C) causing the performance monitoring item updating means to generate or update the performance monitoring item change table started by the job starting means and the job ending means; and (d) causing the performance information collecting means to execute the performance monitoring. Referring to a monitoring default value table and the performance monitoring item change table to determine an item for which performance information is to be collected; and (e) causing the timer to activate the performance information collecting means at regular time intervals. And (f) the monitoring identification flag automatic updating means refers to the contents of the performance information storage file and updates the job JCL in the JCL storage file according to designation from the second operation terminal. Recording a program for causing a computer to execute the steps of adding and deleting an item identification flag.

[0018]

Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the present invention. Referring to FIG.
The embodiment of the present invention includes a JCL (job control language) storage file 10, a job JCL 20, a computer system 30, a performance information storage file 40, an operation terminal 50, and an operation terminal 60.

The business program input to the computer system 30 is expressed in a specific description language.
It is described in L (job control language) format and stored in the JCL storage file 10. JCL storage file 10
Is a file created in a storage device such as a magnetic disk device or an electronic disk device. When executing the business program, the operator uses the operation terminal 60 to instruct the computer system 30 to execute. Job JCL2
0 describes a procedure for executing the load module, and is one business program input to the system one by one. The monitoring item identification flag 201 is described in the job JCL 20 and indicates information on which performance information is to be monitored.

A job is a job starting means 301.
Is a management unit in the case where the business program of the job JCL 20 is executed in the computer system 30 by the execution of the processing by.

The computer system 30 includes the storage device 3
1. Job start unit 301, performance monitoring item update unit 30
2. Job ending means 303, performance information collecting means 304,
Timer 305 and monitoring identification flag automatic updating means 30
6 inclusive.

The storage device 31 stores the computer system 3
Although it is a memory in 0, it may be a file created on an electronic disk device or a magnetic disk device. Further, the storage device 31 includes a management table 311, a performance monitoring item change table 312, and a performance monitoring default value table 313.

The job start means 301 is means (for example, a program) for analyzing the job JCL 20 and starting the execution of the program when a business program is input to the computer system 30. Hereinafter, the term “job” is used as a unit for managing a business program in the computer system 30.

The job ending means 303 is used when the execution of the business program in the computer system 30 ends.
This is means (for example, a program) for releasing resources used in the computer system 30.

The management table 311 is a table created in the storage device 31 and storing control information for each job. Performance monitoring item updating means 302
Are job start means 301 and job end means 303
And a means (for example, a program) for updating the performance monitoring item change table 312. The performance monitoring item change table 312 stores the performance monitoring item updating unit 3.
02 is a table created in the storage device 31 and holds information on the types of performance monitoring items to be monitored that change with job fluctuations.

The performance monitoring default value table 313 is a table created in advance in the storage device 31 in the computer system 30 and holds information on the default value of the type of performance information that the computer system 30 should always collect. I have.

The performance information collecting means 304 includes a performance monitoring default value table 313 and a performance monitoring item change table 31.
2 is a means (for example, a program) for determining an item for collecting the performance information of the computer system 30 based on the logical sum of the information collection designation of both tables and collecting the information.

The timer 305 is connected to the computer system 3
This is an interval timer for activating the performance information collecting means 304 at regular time intervals within 0.

The performance information storage file 40 is a file created in a storage device such as a magnetic disk device or an electronic disk device in order to store and hold the information collected by the performance information collecting means 304 in a record of a certain format. is there.

The monitoring identification flag automatic updating means 306 updates the job JCL 20 in the JCL storage file 10 and designates the monitoring item identification flag 201 by referring to the contents of the performance information storage file 40 and specifying from the operation terminal 50. This is means (for example, a program) for performing processing of deleting or deleting.

Next, the operation of the embodiment of the present invention will be described with reference to the drawings. Figure 2 shows JCL storage file 1
Job J including the monitoring item identification flag 201 existing at 0
It is explanatory drawing which shows the example of CL20.

The description in the row 2002 in FIG. 2 is an example of the monitoring item identification flag 201. In this case, the CP for monitoring the CPU (not shown) in the computer system 30 is used.
There is a description U and a description FIO that monitors the number of I / Os (not shown) for each file. By this designation, it is understood that monitoring of the CPU and monitoring of the number of I / Os for each file are requested during execution of the business program.

By specifying parameters for items that can be monitored in the computer system 30, it is possible to set which performance item the business program requests to monitor. The items to be monitored are not the resources of the job itself but are monitored by the entire computer system 30. If FIO is written in line 2002 in FIG. 2,
This indicates that the number of I / Os for all files used by the program running on the computer system 30 will be monitored.

In this example, JCL is used as a language for describing the execution of the business program. However, even when the execution description language is other than JCL, the same operation is performed by adding the monitoring item identification flag 201 in the description language of the business program. It can be carried out. Further, in this example, the load execution unit is a job of the business program. However, it is also possible to control the execution in steps or in units of processes which are the execution of the program in the computer system 30.

Next, the operation of the job start means 301 will be described with reference to FIG. FIG. 3 is a flowchart showing the operation of the job start means 301. The business program described in the job JCL 20 is input to the computer system 30 and starts execution with the description of the monitoring item identification flag 201, and the start processing is performed by the job start means 301. The process in the job start unit 301 is the first process that is always executed when requesting the computer system 30 to execute a business program.

The job program described by the job JCL 20 input to the computer system 30 is first analyzed in step 32, and the job JCL 20 is analyzed.
An analysis result for determining whether the monitoring item identification flag 201 is described in 20 is created. In step 33, the analysis result of the job JCL 20 in step 32 is determined. If there is an unanalyzable error (error) in the job JCL 20, the processing is interrupted and the execution of the corresponding business program is rejected. If there is no error in the job JCL 20, the process of the next step 34 is executed.

In step 34, the result of the analysis of the job JCL 20 is determined, and if the monitoring item identification flag 201 is described, step 35 for executing the processing by the performance monitoring item updating means 302 is executed. Next, in step 36, the contents of the monitoring item identification flag 201 are copied as management information to an area created for each job in the management table 311 in the storage device 31. The analysis result of the job JCL 20 is determined, and if there is no description of the monitoring item identification flag 201, the process ends without performing the processing in steps 35 and 36.

Next, the management table 311 will be described with reference to FIG. FIG. 4 is an explanatory diagram showing the contents of the management table 311. Referring to FIG. 4, the management table 311 includes a job occurrence number that increases by one each time a job is generated, the name of the user who executed the job, the account name that is the name of the charging group that executed the job, and the name of the job. A job name and the contents of the monitoring item identification flag 201 in the job JCL 20 that need to be set are included. In the example of FIG. 4, “CPU” indicating a CPU monitoring request and “FIO” indicating a monitoring request of the I / O count for each file are included.

Next, the operation of the job ending means 303 will be described with reference to FIG. FIG. 5 shows job ending means 3
11 is a flowchart showing the operation of the third embodiment. The job ending unit 303 is a process executed once at the end when a job being executed in the computer system 30 is ended. In step 401, it is determined whether or not the monitoring item identification flag 201 is specified for this job by referring to the management table 311 in which the contents of the monitoring item identification flag 201 are copied. At the end of the job for which the monitoring item identification flag 201 is specified, in step 403 the performance monitoring item updating unit 30
2 is executed. When the job for which the monitoring item identification flag 201 is not specified ends, the process ends without executing step 403.

In step 404, the job ending means 303 stores the execution history information of the job to be ended in the performance information storage file 40. FIG. 11 is an explanatory diagram showing the contents of job execution history information. FIG.
, The execution history information includes an information type for identifying a record, a job occurrence number that increases by one each time a job is generated, a user name that is the name of the user who executed the job, and a billing amount that executed the job. It includes an account name as a group name, a job name as a job name, an execution start time, and an execution end time.

Next, the operation of the performance monitoring item updating means 302 will be described with reference to FIG. FIG. 6 is a flowchart showing the operation of the performance monitoring item updating means 302.
In step 501, it is determined whether the process of step 35 in FIG. 3 has been executed by the job start unit 301, or the process of step 403 in FIG.
The performance monitoring item updating means 3 determines whether or not the process has been executed.
02. When the process of step 35 is executed, step 502 is executed next. If the processing of step 403 has been executed, then step 5
08 is executed.

First, the processing when step 35 is executed will be described. In step 502, the job JC
One content of the monitoring item identification flag 201 described in L20 is extracted. In step 503, the performance monitoring item change table 312 in the storage device 31 is searched using the content of the one monitoring item identification flag 201 extracted as a key. FIG. 7 is an explanatory diagram showing the format of the performance monitoring item change table 312. Referring to FIG. 7, the performance monitoring item change table 312 is a table in which each monitoring item 601 has items of an item name 602, a setting 603, and a requested job number 604, and the item name 602 is used as a key. The setting 603 takes two values, “ON” and “OFF”, and when the value of the number of requested jobs 604 is 1 or more, it becomes “ON”, meaning that there is a monitoring request. The number of requested jobs is 60
When the value of 4 is 0, it becomes “OFF”, which means that there is no monitoring request.

The example of the table shown in FIG. 7 will be described in detail.
Since the number of requested jobs 604 of the CPU monitoring request is “10”, the setting 603 is “ON”. Similarly, the setting 603 of the memory monitoring request is “OFF” because the number of requested jobs 604 is “0”. Step 50
3, the performance monitoring item change table 312 is searched, and in step 504, the monitoring item identification flag 2 is searched.
If the content of the item name 602 that matches the specified content of 01 is detected, step 505 is executed.

If there is no match, step 50
In step 6, a new entry is created in the performance monitoring item change table 312, a new item name is added to the item name 602, the setting 603 is set to “ON”, and the number of requested jobs 604 is set to “1”.

In step 505, the setting 603 of the corresponding item name 602 in the performance monitoring item change table 312 is set to "ON", and the value of the requested job number 604 is incremented by one.
In step 507, it is determined whether the processed item is the last specified item. If not, the process returns to step 502, and if it is the last, the process ends.

Next, the processing when step 403 is executed will be described. In step 508, the management table 311 of the job to be ended is searched, and one item corresponding to the monitoring item identification flag 201 is extracted. Step 50
In step 9, the performance monitoring item change table 312 is searched by using the designated item of the extracted flag as a key. As a result of the search, in step 510, “1” is subtracted from the number of required jobs 604 corresponding to the corresponding monitoring item 601. If the number of required jobs 604 becomes “0” after the operation, the setting is set to “OFF”. To "." In step 511, if the processed monitoring item identification flag 201 is the last one, the processing is terminated. If not, control is returned to step 508.

FIG. 8 is an explanatory diagram showing the format of the performance monitoring default value table 313. Referring to FIG. 8, the performance monitoring default value table 313 has an item name 702 and a setting 703 for each default value setting item 701. Entries are created in the performance monitoring default value table 313 for all performance monitoring items that can be collected by the computer system 30. The item name 702 is a name for identifying an item for performance monitoring, and the monitoring item identification flag 20
It corresponds to the contents of 1. The settings 703 are “ON” and “O”
FF "and the corresponding default value setting item 70
When information 1 is to be collected, it is set to “ON”, and when it is not desired to be collected, it is set to “OFF”.

The example of FIG. 8 will be described in detail. In the CPU monitoring request, the item name 702 is “CPU”, and the setting 703 is “ON” to collect this information.
Also, the file I / O monitoring request has an item name 702 of “F
The setting 703 is “OFF” because the user does not want to collect this information in “IO”. If the amount of data to be collected at a time is small or basic performance data as in the CPU information of this example, the data should always be collected.
The setting 703 should be set to “ON”. Conversely, for information such as a file I / O monitoring request, for which detailed performance data can be obtained but the amount of data is extremely large, or data that should be collected only when a special function is used, refer to “OF
It is desirable for operation to set it to "F".

Next, the operation of the performance information collecting means 304 will be described with reference to FIGS. FIG. 9 is a flowchart showing the operation of the performance information collecting unit 304. The performance information collecting means 304 is activated upon being notified by the timer 305 at regular intervals. The activated performance information collecting unit 304 collects information on the performance of the entire system of the computer system 30 according to the specification of the collection item.

The detailed processing is as follows. In step 801, one default value setting item 701 is read from the performance monitoring default value table 313. Step 802
Then, the default value setting item 70 read in step 801
1 is determined, and the value of the setting 703 is set to “OF”.
In the case of "F" (there is no setting), the processing of step 803 is performed. Conversely, if the value of the setting 703 is “ON”, the process of step 805 is performed because performance monitoring is performed.

In step 803, an item that is equal to the item name 702 of the default value setting item 701 read in step 801 is read from the performance monitoring item change table 312. In step 804, the corresponding item exists in the performance monitoring item change table 312, and the setting 603 is “ON”.
If it is, step 805 is executed. If it is “OFF”, step 805 is not executed and step 806 is executed.
Execute In step 805, the corresponding performance information is collected, a record thereof is created, and the record is stored in the performance information storage file 40. In step 806, if the data extracted from the performance monitoring default value table 313 is not the last, the control is returned to step 801. If the data is the last, the process ends.

In the case of the examples of FIGS. 7 and 8, when the processing by the performance information collecting means 304 is executed, a “CPU monitoring request”, a “memory monitoring request”, and a “file I / O” are executed according to the flowchart of FIG. The “monitoring request” is executed by the performance information collecting unit 304. "Job unit CPU
Since the setting 603 of “monitoring request” is “OFF” and the setting 703 of “job unit CPU monitoring request” is also “OFF”, performance information is not collected. In the example of FIG. 7 and FIG.
Although only the items have performance monitoring targets, the same processing is performed when there are more items.

As described above, the item for collecting the performance information of the computer system 30 is the performance monitoring default value table 31.
3 and the monitoring request of the job being executed. The data collected in this way and stored in the performance information storage file 40 is used as data for a system administrator to perform performance monitoring.

Next, the operation of the monitoring identification flag automatic updating means 306 will be described with reference to FIGS. FIG.
0 is a flowchart showing the operation of the monitoring identification flag automatic updating means 306. Monitoring identification flag automatic updating means 3
06 refers to the contents of the performance information storage file 40 according to an instruction from the operation terminal 50, and refers to the JCL storage file 10
(For example, a program) for adding or deleting the monitoring item identification flag 201 of the job JCL 20 in the
It is. Monitoring item identification flag 201 in job JCL 20
Can be written individually using a normal editor, or the conditions for adding to the job JCL 20 are determined, and the JC
When the information for determining L is stored, the monitoring identification flag automatic updating means 306 can be used.

Referring to FIG. 10, first, in step 901, the monitoring identification flag automatic updating means 306 inputs a condition and a change content for updating the JCL from the operation terminal 50. The conditions and changes to be input are, for example, I / O
When a performance problem occurs as a bottleneck, specify the time zone in which the problem occurred, and add a monitoring item identification flag 201 for monitoring I / O to the JCL of the job executed during that time. and so on.

In step 902, the job name, user name, account name, and the like of the JCL that matches the conditions are extracted with reference to the job execution history information in the performance information storage file 40. However, in order to execute step 902, the condition is that operation history information for each job is stored as performance information in the performance information storage file 40. In step 903, a JCL list table that meets the conditions is created. In step 904, step 90
3, a job matching the condition is found in the JCL storage file 10 and the job JCL is searched.
20 is added with the designated monitoring item identification flag 201. Thus, the monitoring item identification flag 201 can be automatically added.

Further, a performance information collecting method including the following steps (a) to (f) can be realized.

(A) causing the job starting means 301 to analyze the job JCL 20 as the business program stored in the JCL storage file 10 from the operation terminal 60; and (b) causing the job ending means 303 to execute the business program. To release the used resources when the process is completed; and (c) performance monitoring item updating means 302
The job start means 301 and the job end means 3
03, the step of generating or updating the performance monitoring item change table 312, and (d) causing the performance information collecting unit 304 to refer to the performance monitoring default value table 313 and the performance monitoring item change table 312 to perform performance information. Determining the items to be collected, (e) causing the timer 305 to start the performance information collecting means 304 at regular time intervals, and (f) storing the performance information storage file in the monitoring identification flag automatic updating means 306. A step of referring to the contents of 40 and updating the job JCL 20 in the JCL storage file 10 and adding / deleting the monitoring item identification flag 201 by designation from the operation terminal 50.

Further, as shown in FIG.
A recording medium 120 that records a program that causes the computer 100 to execute the step (f) can be realized.

[0060]

The first effect is that the collection of performance information can be selectively output only during the valid time zone.

The reason is that in the present invention, when a load is executed, which performance collection is automatically requested in the system is registered in the system, and only during the execution period is the corresponding information collected by the performance information collection means of the computer system. Because it is automatically collected.

A second effect is that performance information that is more effective than performance information stored in the same file area as before can be stored.

The reason for this is that, by setting no monitoring items for non-essential business programs, the performance monitoring items can be reduced and the amount of information for performance monitoring can be automatically reduced during execution of those businesses. Because.

A third effect is that the load collection items can be automatically set.

The reason is that, when a system manager identifies a business program that may have caused the problem, for example, when a performance problem occurs in the computer system, many business programs that may be the cause of the problem are identified. This is because, in many cases, performance monitoring items can be collected without fail by a reproduction test or the like by automatically adding a monitoring item identification flag to the business programs as candidates.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is an explanatory diagram illustrating an example of a job JCL.

FIG. 3 is a flowchart illustrating an operation of a job start unit.

FIG. 4 is an explanatory diagram showing contents of a management table.

FIG. 5 is a flowchart illustrating an operation of a job ending unit.

FIG. 6 is a flowchart illustrating an operation of a performance monitoring item updating unit.

FIG. 7 is an explanatory diagram showing a format of a performance monitoring item change table.

FIG. 8 is an explanatory diagram showing a format of a performance monitoring default value table.

FIG. 9 is a flowchart illustrating an operation of a performance information collecting unit.

FIG. 10 is a flowchart showing an operation of a monitoring identification flag automatic updating unit.

FIG. 11 is an explanatory diagram showing the contents of execution history information.

FIG. 12 is a block diagram showing a computer and a recording medium.

[Description of Signs] 10 JCL storage file 20 Job JCL 30 Computer system 31 Storage device 40 Performance information storage file 50 Operation terminal 60 Operation terminal 100 Computer 120 Recording medium 201 Monitoring item identification flag 301 Job start means 302 Performance monitoring item update means 303 Job ending means 304 Performance information collecting means 305 Timer 306 Monitoring identification flag automatic updating means 311 Management table 312 Performance monitoring item change table 313 Performance monitoring default value table

Claims (10)

[Claims] An information processing apparatus includes a JCL storage file for storing a job JCL, a computer system, a performance information storage file, a first operation terminal, and a second operation terminal, wherein the job JCL stores performance information. The computer system includes a storage device, a job start unit, a performance monitor item update unit, a job end unit, a performance information collection unit, a timer, and a monitor identification flag automatic update unit. A management table that stores control information for each business program, a performance monitoring item change table that stores information on the type of performance monitoring item to be monitored that changes with the execution of the business program, and A default value table that indicates the type of performance information to be collected
(A) the job start means analyzes the job JCL as a business program stored in the JCL storage file from the first operation terminal; and (b) the job end means executes the business program. At the end,
Releasing the used resources; (c) the performance monitoring item update means is activated by the job starting means and the job ending means to generate or update the performance monitoring item change table; and (d) the performance monitoring item change table. The information collecting means determines an item for collecting performance information with reference to the performance monitoring default value table and the performance monitoring item change table, and (e) the timer sets the performance information collection at regular time intervals. (F) The monitoring identification flag automatic updating means refers to the contents of the performance information storage file and updates the job JCL in the JCL storage file by designation from the second operation terminal. A performance information collection device for adding and deleting the monitoring item identification flag. 2. The job start means analyzes the contents of the business program, determines whether the monitoring item identification flag is described in the job JCL, and outputs an unanalyzable error to the job JCL. In some cases, the processing is interrupted. When there is no error, if there is a description of the monitoring item identification flag, the performance monitoring item updating means executes the processing. 2. The performance information collection device according to claim 1, wherein the content of the monitoring item identification flag is copied as management information to the management table in a storage device. 3. The job ending unit refers to the management table for the business program to be ended, determines whether the monitoring item identification flag is specified in the business program, and determines whether the monitoring item identification flag is specified. 2. The performance monitoring item updating means executes processing at the end of the business program.
Or the performance information collection device described in 2. 4. The performance information according to claim 1, wherein the performance monitoring item change table includes an item name, a setting, and the number of requested jobs for each monitoring item for which performance information is collected. Sampling equipment. 5. The performance monitoring item updating means determines whether processing by the job starting means is being executed or whether processing by the job ending means is being executed. Is being executed, one content of the monitoring item identification flag described in the job JCL of the business program is taken out, and the content of the taken out monitoring item identification flag is used as a key, and The performance monitoring item change table is searched, and when an item name that matches the specified content of the monitoring item identification flag is detected, “1” is added to the requested job number of the corresponding item name in the monitoring item change table. If a matching item name is not detected, a new entry is created in the performance monitoring item change table, and the new item name is added to the item name. Is set to "ON", the number of requested jobs is set to "1", and it is determined whether the processed item is the last designated item. If not, the content of the monitoring item identification flag is set to 1 When the processing is completed, the processing is ended. (B) When the processing by the job ending means is being executed, the management table of the business program to be ended is searched. One monitoring item corresponding to the monitoring item identification flag is fetched, the performance monitoring item change table is searched using the fetched monitoring item as a key, and "1" is subtracted from the number of requested jobs corresponding to the monitoring item. If the number of requested jobs has become “0”, the setting is changed to “OFF”, and if the processed monitoring item identification flag is the last one, the process is terminated. Watch terms Performance information collection apparatus according to claim 4, wherein the back monitoring items corresponding to the identification flag to one taken out process. 6. The performance monitoring default value table includes an item name that is a name for identifying an item for performance monitoring corresponding to the content of the monitoring item identification flag for each default value setting item, and “ON”. "" And "OFF". If you want to collect information on the corresponding default value setting item,
6. The performance information collection device according to claim 5, further comprising a setting of "N" and "OFF" when the collection is not desired. 7. The performance information collecting means is activated upon being notified by the timer at regular intervals, reads one of the predetermined value setting items from the performance monitoring default value table, and stores the corresponding setting value. (A) When the value of the setting in the performance monitoring default value table is “OFF”, an item name equal to the item name corresponding to the read default value setting item is replaced with the performance monitoring item. If the corresponding item name is read from the change table and the corresponding item name exists in the performance monitoring item change table and the setting in the performance monitoring item change table is “ON”, the performance information corresponding to the item name is read. Collected and stored in the performance information storage file. (B) If the value of the setting in the performance monitoring default value table is "ON", the item name in the performance monitoring default value table The corresponding performance information is collected and stored in the performance information storage file. (C) If the default value setting item extracted from the performance monitoring default value table is not the last, the performance monitoring default value table 7. The performance information collection device according to claim 6, wherein the process returns to the process of reading one default value setting item and ends if it is the last. 8. A monitoring identification flag automatic updating unit inputs a condition and a change content for updating the JCL from the second operation terminal, and refers to performance information information of the business program in the performance information storage file. 8. The performance information according to claim 7, wherein the business program matching the condition is detected, and a monitoring item identification flag specified by the change content is added to the job JCL corresponding to the business program. Sampling equipment. 9. It has a JCL storage file for storing a job JCL, a computer system, a performance information storage file, a first operation terminal, and a second operation terminal, wherein the job JCL stores the performance information. The computer system includes a storage device, a job start unit, a performance monitor item update unit, a job end unit, a performance information collection unit, a timer, and a monitor identification flag automatic update unit. A management table that stores control information for each business program, a performance monitoring item change table that stores information on the type of performance monitoring item to be monitored that changes with the execution of the business program, and Default value indicating the type of performance information to be collected Performance for performance information collection device including performance monitoring default value table Information collection method,
(A) causing the job start means to analyze the job JCL as a business program stored in the JCL storage file from the first operation terminal;
(B) causing the job ending means to release used resources when the execution of the business program ends; and (c) causing the performance monitoring item updating means to execute the job starting means and the job. Started from the end means,
Generating or updating the performance monitoring item change table; and (d) determining, by referring to the performance monitoring default value table and the performance monitoring item change table, the performance information collecting means to collect performance information. (E) causing the timer to activate the performance information collecting means at regular time intervals;
(F) referring to the content of the performance information storage file, and updating the job JCL in the JCL storage file by referring to the contents of the performance information storage file to the monitoring identification flag automatic updating means, and identifying the monitoring item identification; A step of adding and deleting a flag. 10. A JCL storage file for storing a job JCL, a computer system, a performance information storage file, a first operation terminal, and a second operation terminal,
The job JCL includes a monitoring item identification flag indicating a type of performance information to be collected, and the computer system includes a storage device, a job starting unit, a performance monitoring item updating unit, a job ending unit, a performance information collecting unit, a timer, And a monitoring identification flag automatic updating means, wherein the storage device stores a management table for storing control information for each business program, and information relating to a type of a performance monitoring item to be monitored which changes with the execution of the business program. (A) for the performance information collection device including a performance monitoring item change table and a default value performance monitoring default value table indicating a type of performance information to be constantly collected,
Analyzing the job JCL as the business program stored in the JCL storage file from the first operation terminal; and (b) using the job ending means when the execution of the business program ends. (C) causing the performance monitoring item updating means to be activated or activated by the job starting means and the job ending means to generate or update the performance monitoring item change table; and (d) (E) causing the performance information collection means to determine an item for which performance information is to be collected by referring to the performance monitoring default value table and the performance monitoring item change table;
Activating the performance information collecting means at regular time intervals; and (f) instructing the monitoring identification flag automatic updating means to refer to the contents of the performance information storage file and to execute the second
Recording a program for causing a computer to execute the steps of: updating the job JCL in the JCL storage file and adding / deleting the monitoring item identification flag in accordance with designation from the operation terminal of Medium.