Help

 

Program CPU Usage

The Program CPU Usage report shows the distribution of CPU time used by each active control section of each module in the target program or subsystem.

CPU time used by each control section is given as a percentage of all CPU time used in the measured job step; therefore the CPU time percentages used by all listed control sections total 100.

The report groups all system service routines by function under appropriately named pseudo-sections within the pseudo-module .SYSTEM. Compiler library routines included in the target program or subsystem load modules are grouped by library under a suitably named pseudo-section. For example, COBOL library routines will appear as section name .COBLIB. The module name will be .SYSTEM if the application was compiled with RES.

All user-coded routines appear within the pseudo module .USER. Compiler library routines compiled with the NORES option appear with the module name under this pseudo-module heading.

  • If the control section was not indexed, the report shows activity within a code block, an area whose size in bytes is equal to the report resolution specified when creating the Performance Profile. Each code block is identified by its starting location and its length.

  • If the program was indexed, the report shows activity within procedure boundaries dictated by the programming language.

Exception: Source support is not provided for those application programs that have not reached the Detail Baseline or where the source is not provided.

The presentation uses a tree map method for displaying data.

Report details

Select a resolution from the Resolution drop-down list box to adjust the view of the Program CPU Usage report. Changing the value of the code block in bytes will adjust the resolution of the report. A large code block size allows for a more generalized view of CPU activity. A small code block size allows for a more specific view of CPU activity. The code block default is 2 bytes. An adjusted code block will remain as such until the session ends or another value is selected.

  • Pseudo-Module or Pseudo-Section is the name of the active pseudo-entity.

  • Description is a short description of the function of the pseudo-entity.

  • CPU % Combined shows the sum of CPU % Attrib and CPU % Total. This adds the values in the Attributed column with the values in the Total Column.

  • CPU % Attributed shows the CPU % total of all system services invoked on behalf of the line of the entity. This column shows the CPU activity that resulted in an overhead routine being invoked and is attributed by Strobe back to this line of code. For example, a line with INSPECT may invoke the IGZCINS1 routine, this routine will show up in this column.

  • CPU % Execution shows the complete amount of CPU consumed by the item in question.

Click next to the pseudo-entity to view additional detail:

  • Module shows the name of the active module.

Notes:

    • For .USER programs that execute in a CICS environment, click to export a fix request file that can be imported into the Topaz Workbench to address a performance problem identified by iStrobe.

    • For all module or section names that exceed eight characters, Strobe generates a token, which is an eight-byte identifier. The token comprises the first four characters of the module or section name followed by a hyphen (-) and then the last three characters of the name. Refer to the Token Long Name Cross Reference report to reconcile all tokens with their long names.

  • RMODE indicates whether the module must reside beneath the 16M line (24) or may reside above it (ANY). If the module is a Generalized Object File Format (GOFF) split-RMODE module, a value of SPLIT appears in this column.

  • Description is a short description of the function of the module.

  • Interval Length shows the length, in bytes, of the area covered by the entry line, expressed in decimal notation.

Click next to the Module to view additional detail:

  • Section is the name of the active control section.

  • Language shows the language in which the measured program was written.

  • Description is a short description of the function of the section

Click next to the Section to view additional detail. The additional information presented varies based on the environment of the measured application, whether the code was indexed, and how the Performance Profile was created.

Start of codeblock interval and Source Listing text at start of codeblock

  • Line number shows the line number the compiler assigned to a line in the source program, for indexed programs.

  • Source Text shows statement text associated with the line number, for indexed programs.  [MORE]

  • Location shows the first storage location of the area, relative to the beginning of the control section, expressed in hexadecimal notation.

  • Interval Length shows the length, in bytes, of the area covered by the entry line, expressed in decimal notation.

Click next to the line number to view Services invoked by the location identified in the previous level.  [MORE]

Analysis tips

Click an underlined module name to see an expanded description and performance improvement hints.

Related Topics

Tree Map Reports