GUIDELINE 4610.02 

DISC
Bureau of Information Systems

Effective Date
08/15/2002

Review Date
08/2002

 

1.0 SUBJECT: Compuware – STROBE MVS for Sysplex Monitoring Facility

 2.0 DISTRIBUTION: All Current Users

3.0 FROM: Bruce Roberts, Director of DISC

4.0 PURPOSE: The purpose of STROBE is to analyze and collect performance data for batch and on-line programs written in Assembler, higher level languages like COBOL and FORTRAN, or even 4th generation languages like AS and NATURAL. STROBE can also be effective in analyzing programs that access commercial data bases such as CICS, COOL:GEN, ADABAS and DB2.

5.0 BACKGROUND: STROBE can be used to help evaluate programs which are potentially inefficient. Once these programs are identified, STROBE can then be used to identify lines of code which can be re-written to make more efficient use of computer resources.

The STROBE Administrator can submit, via TSL, measurement requests before the execution of the target program. When the target program begins, STROBE recognizes it and activates a measurement task within the address space of the program. STROBE can also measure programs that are already executing. Interactive panels available on TSO may be used to control STROBE requests which have been previously submitted.

6.0 PROCEDURE: STROBE can be started by using TSO ISPF/PDF or a batch interface. STROBE is menu driven. You should base the selection of programs to be measured by STROBE on either inadequate response time, high CPU time consumption, long batch turn-around, or any other agency criteria. STROBE can monitor programs that are actively running, such as a TSO user or batch job, or a request can be submitted to monitor TSO userids or jobs that will be released or submitted at a later time.

6.1 STROBE has a reporting facility which analyzes the data created by the monitoring of a selected JOB, Started Task, or TSO session. Once the measurement sample has been obtained, the source code of the program can be analyzed and concatenated with the execution sample data set during the reporting phase, which will display the actual lines of source code indicating their relative CPU usage on a histogram.

6.2 STROBE will show Run Time, CPU Time, Wait Time and Total Time of the programs that are monitored. STROBE will list all modules individually. Additionally, there are pseudo-entities that will include such things as group openings, closings, (file management operations), SVC’s used VTAM usage, waits, etc.

6.3 STROBE can produce the following reports:

    6.3.1 Measurement Session Data Reports:

      1. Job environment

      2. Measurement parameters

      3. Measurement statistics

      4. Report parameters

      6.3.2 Time Distribution of Activity Level Report: Displays a chronological record of the level of activity of the program or sub-system and each file access activity performed under its control.

      6.3.3 Resource Demand Distribution Report: Summarizes the use of CPU and I/O resources by the task execution and file access activities.

      6.3.4 Wait Time by Module Report: Shows all the modules, control sections, pseudo-modules and pseudo-sections in which the target program was found to be in a wait state.

      6.3.5 Data Set Characteristics Report: Shows the characteristics of all data sets accessed for the measured job step during the measurement session.

      6.3.6 I/O Facility Utilization Summary Report: Shows by device and volume the percentage of run time used by the CPU accessing each data set.

      6.3.7 Most Intensively Executed Procedures Report: Lists the ten most used procedures in the programs executed in the measured job step.

      6.3.8 Most Extensive Inactive Areas Report: Shows the ten largest areas of main storage that never appear to be accessed.

      6.3.9 Program Section Usage Summary Report: Shows the distribution of CPU time used by each active control section for each module in the target program or sub-system.

      6.3.10 Transaction Usage Summary Report: Shows the distribution of CPU usage among transactions processed by the target sub-system such as CICS.

      6.3.11 Program Usage by Procedure Report: Details the time the CPU spent executing code within each area of each control section for each module of the program.

      6.3.12 Transaction Usage by Control Section Report: Lists transactions by identification code (tran-id), module name, and control section name.

      6.3.13 DASD Usage by Cylinder Report: A detailed analysis of the time that the CPU spent accessing each cylinder of each direct access device during execution of the target program.

      6.3.14 Attribution Report: Identify sites of invocation of selected service routines.

      6.4.15 Data Set Characteristics Supplement Report: Shows the OPEN INTENT/PROCESSING MODE, which indicates whether the application opens files for input, output, or both. For VSAM files it shows records are processed sequentially or directly.

      6.4.16 VSAM LSR Pool Statistics Report: Provides resource information for Data Sets allocated to local shared resources.

      6.4.17 Subsystem-Specific Report: Provides CPU wait time information and consumption information.

      6.4.18 Coupling Facility Activity Report: Identifies system-wide coupling facility activity.

7.0 HISTORY: This Guideline was originally issued #4610.00 dated 11/01/92. Re-issued #4610.01, 01/04/2002. Re-issued #4610.02, 08/09/2002.

8.0 CONTACT PERSON: Deputy Director, Bureau of Information Systems
785-296-3343