Strobe Performance Profile for IMS Supervisory Regions

This topic shows reports from two sample Strobe Performance Profiles: one from an IMS control region and one from a DL/I Separate Address Space (DLISAS).

Analyzing the Reports from an IMS Control Region

The Measurement Session Data report (see the following figure) describes the environment during a measurement session. When Strobe measures a job step executing in IMS, the report shows in the SUBSYSTEM field the version and release number of IMS, a region identifier, and the Local Storage Option (LSO). In this example, the region identifier CTL shows that this is an IMS control region. (For a list of other region identifiers, see Analyzing-the-Reports-for-an-MPR-Region.)

Measurement Session Data Report

Choosing Between Execution and I/O Reports

To examine CPU activity, analyze the transaction reports to determine which function-specific transactions used the most CPU time. Analyze the Program Module reports to determine which modules used the most CPU time, as shown in Identifying CPU Usage in a Control Region.

To examine I/O activity, analyze the Resource Demand Distribution, Data Set Characteristics, I/O Facility Utilization Summary, and DASD Usage by Cylinder reports, as shown in Identifying I/O Activity in a Control Region.

The Transaction Summary report (Transaction Summary Report) shows the distribution of CPU time among function-specific pseudo-transactions within the region. Each pseudo-transaction appearing in the report is described in more detail in the Transaction Usage by Control Section report.

Identifying CPU Usage in a Control Region

In the transaction reports for control regions, Strobe assigns CPU time to pseudo-transactions representing the IMS ITASK being executed. The Strobe-assigned pseudo-transaction names are made up of a three-character task identifier followed by a slash (/), concatenated with characters four through seven of the active module name. (For a list of common ITASK identifiers, see Transaction Identification for IMS System Regions.)

For example, the name of the load module that performs database recovery management is DFSDSST0. To create its pseudo-transaction, Strobe begins with the identifier CTL/ to indicate that the task is the control region task, and appends the characters four through seven of the load module name to produce the pseudo-transaction name “CTL/DSST”.

If Strobe cannot determine an ITASK identifier, it assigns as the pseudo-transaction name the name of the load module first entered by the task. In the example in the following figure, the DB Interval Timer task starts with load module DFSDFLD0 and is therefore assigned a pseudo-transaction name of DFSDFLD0.

Transaction Summary Report

The Transaction Usage by Control Section report (see the following figure) shows, for each IMS pseudo-transaction, the CPU time spent in each module executed to perform the service function represented by the pseudo-transaction.

Activity in DFS and DBF modules is compressed as .IMS. The true name of each module appears in the SECTION NAME field of the report detail lines and is followed by a brief description of its function.

Transaction Usage by Control Section Report

The Program Section Usage Summary report (see the following figure) shows the activity of each pseudo-module that was active during the measurement session. In this example, most of the CPU activity is used by IMS supervisory functions.

Program Section Usage Summary Report

The Program Usage by Procedure report (see the following figure) shows the actual names of all modules within the pseudo-modules. If control section mapping data was available the report shows control sections within modules. Strobe also displays function descriptors for the control section or module.

In this example, modules DFSDSST0, DFSDPDM0, and DFSDBAU0 were responsible for over one-third of the CPU time used by IMS control functions.

Program Usage by Procedure Report

Identifying I/O Activity in a Control Region

The Resource Demand Distribution report (see the following figure) summarizes the use of CPU and I/O resources by the task execution and file access activities occurring in the control region. All percentages of resource use are based on the total run time.

In this example, DDNAME DFSWADS0 (the Write-Ahead Data Set) was responsible for the most I/O activity. For more details about this data set, examine the Data Set Characteristics report.

Resource Demand Distribution Report

The Data Set Characteristics report (see the following figure) shows, for each data set accessed during the measurement session, the access method, block size, and data set name. The number of I/O operations (EXCPs) performed during the measurement session appears as well. In the example shown, the EXCP counts confirm the heavy use of the Write-Ahead Data Sets (WADS) identified in the Resource Demand Distribution report. To determine what volume this data set resides on, examine the I/O Facility Utilization Summary report.

Data Set Characteristics Report

The I/O Facility Utilization Summary report (see the following figure) summarizes I/O access by device, volume, and ddname. In this example, the Write-Ahead Data Set resides on a single unit. For more information about I/O activity for each cylinder, examine the DASD Usage by Cylinder reports.

I/O Facility Utilization Summary Report

The DASD Usage by Cylinder report (see the following figure) identifies data set activity by unit, volume, ddname and cylinder. This example shows the distribution of I/O activity across each cylinder of the WADS.

DASD Usage by Cylinder Report

Since usage of the WADS by transactions is single-threaded, it is important to ensure that the WADS has a large enough allocation to service the transactions that use it the most.

Analyzing the Reports for a DL/I Separate Address Space

The Measurement Session Data report (see the following figure) describes the environment during a measurement session. When Strobe measures a job step executing in IMS, the report shows in the SUBSYSTEM field the version and release number of IMS, a region identifier, and the Local Storage Option (LSO). In this example, the region identifier DLS shows that this is a DL/I Separate Address Space (DLISAS). (For a list of other region identifiers, see Analyzing-the-Reports-for-an-MPR-Region.)

Measurement Session Data Report

Choosing Between Execution and I/O Reports

Because the DLISAS owns all of the IMS database data sets, it is I/O-intensive. To examine the I/O activity, analyze the Resource Demand Distribution, Data Set Characteristics, I/O Facility Utilization Summary, and DASD Usage by Cylinder reports, as shown in Identifying I/O Activity in a DLISAS.

To examine the CPU activity, analyze the Transaction Summary and Transaction Usage by Control Section reports, as shown in Identifying CPU Activity in a DLISAS.

Identifying I/O Activity in a DLISAS

The Resource Demand Distribution report (see the following figure) summarizes the use of CPU and I/O resources by the task execution and file access activities occurring in the DLISAS region. All percentages of resource use are based on the total run time. The SERVICED BY I/O percentages can be greater than 100% because the DLISAS region owns the database data sets and handles multiple concurrent I/O activities for dependent regions.

Resource Demand Distribution Report

In this example, five data sets (CRZDAP01, CWADAP01, CWPDAP01, FC1DAP01, and FC2DAP01) are responsible for most of the I/O activity in the region. For more details about these data sets, examine the Data Set Characteristics report.

The following figure shows an abbreviated version of the Data Set Characteristics report. This example shows, for each data set accessed during the measurement session, the access method, block size, and data set name. The number of I/O operations performed (EXCPs) during the measurement session appears as well. In the example shown, the EXCP counts confirm the heavy use of the data sets identified in the Resource Demand Distribution report. To determine which volumes these data sets reside on, examine the I/O Facility Utilization Summary report.

Data Set Characteristics Report

The I/O Facility Utilization Summary report (see the following figure) summarizes I/O access by device, volume, and ddname. In this example, the five data sets reside on different units, making access more efficient. For more information about I/O activity on each cylinder, examine the DASD Usage by Cylinder reports.

I/O Facility Utilization Summary Report

Identifying CPU Activity in a DLISAS

The Transaction Summary report (see the following figure) shows the distribution of CPU usage among function-specific pseudo-transactions within the region. Each pseudo-transaction appearing in the Transaction Summary report is described in more detail in the Transaction Usage by Control Section report.

In the transaction reports for DLISAS regions, Strobe assigns CPU time to pseudo-transactions representing the IMS ITASK being executed. The

Strobe-assigned pseudo-transaction names are made up of a three-character task identifier followed by a slash (/), concatenated with characters four through seven of the active module name. For a list of common ITASK identifiers, see Transaction Identification for IMS System Regions.

Transaction Summary Report

In this example, transaction DLI/019L VSAM RECORD MANAGEMENT is responsible for most of the CPU time. For more information, examine the Transaction Usage by Control Section report for this transaction.

The Transaction Usage by Control Section report (see the following figure) shows, for each IMS pseudo-transaction, the CPU time spent in each module executed to perform the service function represented by the pseudo-transaction. In this example, all of the CPU time was spent in one VSAM module