Skip to Content

Resource demand distribution report

The Resource Demand Distribution report summarizes the use of CPU and I/O resources in the measured job step. Strobe bases all percentages of resource use on the total run time. The tasks and resources listed are the same as those in the Time Distribution of Activity Level report. The report lists them in order of decreasing solo activity, as indicated by the values shown in the column headed SOLO IN EITHER.

If you specified the NOTASK reporting option, this report contains a single line showing the combined time distribution of all tasks.

Resource Demand Distribution Report

New_RDD_report.png

TASK OR DDNAME

The first entries under TASK OR DDNAME (data definition name) describe task activity within the address space. Subsequent entries describe file access activities. The report lists the activities within each task or ddname in order of decreasing solo activity.

If you specify the ALLTASK parameter, then the report will show all of the tasks. Similarly, if you specify the ALLDD parameter, then the report will list all of the ddnames in which Strobe detected activity. The report is limited to a maximum of 1000 unique tasks. If more than 1000 tasks are seen during the measurement, a warning message, STR2124W, is issued when the profile report is built.

The report identifies tasks by the name of the module in which execution was initiated or by the name specified in the invocation of an ATTACH macro. In addition to the module name, Strobe prints the task control block (TCB) address which uniquely identifies each task. 

The separate entry labeled .FILEMGT is a pseudo-activity, which combines file access operations (such as open and close activities) or other file access activities performed by IOS drivers that Strobe could not assign to a specific file. If their collective activity is sufficiently low, the report will include them in the .OTHER line entry instead. (For more information on pseudo-activities, see Pseudo-Entities.)

RESOURCE

Resources in this column are listed for the following activities:


    • For file access activities, the model number of the I/O device that serviced each task appears. If the model is not available, the following device types appear:

      C

      Communication

      DA

      Direct access

      G

      Graphic

      MT

      Magnetic tape

      UR

      Unit record

    • For CPU activities, the literal CPU appears.
    • For activity in I/O services routines, the literal I/O appears.
    • For files processed by the BatchPipes subsystem, the literal B_PIPES appears.

SERVICED BY CPU

The percentage of run time spent by a CPU in servicing or performing the activity. The sum of CPU service percentages is equal to or greater than the EXEC time percentage reported in the Measurement Session Data report. This percentage may be greater than the EXEC percentage in a multitasking environment when several CPUs simultaneously execute for the measured job step.

The reported CPU service time percentage for execution of a task includes the time required for the program itself, the subroutines it invokes, library routines, and some supervisor calls. It does not include CPU time associated with file management operations assigned to individual file access activities or to .FILEMGT.

The CPU service time percentage reported for a file access activity represents the CPU time required to execute IBM-supplied access method routines in accessing the subject data set. Strobe assigns activity in data management SVCs, some ISAM and BDAM access method modules, and all unidentified files to the pseudo-activity .FILEMGT.

Warning

Important

A high percentage of run time in a CPU resource indicates that a small percentage of time processing an application program involved I/O. With this information, you can refer to the Program-Section-Usage-Summary-report and analyze a specific application program’s resource use.

SERVICED BY I/O

The percentage of run time spent by I/O facilities in servicing the activity. Service time includes all time during which the servicing unit was busy performing the required access operations. For example, service time for disk I/O operations includes time spent by the servicing disk access mechanism during the seek, search, and data transfer operations.

The I/O facility use by each file access activity is further broken down by unit and volume in the I-O-Facility-Utilization-Summary-report. For more information on file access activity, consult that report.

For task execution, the I/O service time percentage is always zero. Service time for access to elements such as program load modules and overlays is shown in STEPLIB or JOBLIB activity.Strobe assigns I/O operations performed simultaneously for several data sets (such as multiple open and close operations) to the pseudo-activity .FILEMGT.

SERVICED BY EITHER

The percentage of run time in which either CPU or I/O facilities were servicing the activity.

SOLO IN CPU

The percentage of run time during which a CPU was servicing the activity and no file access activity was occurring.

For tasks, CPU solo time is the time during which the task was being executed and no other activity was taking place.

For a file access, CPU solo time is the time during which a CPU was executing file management routines for the activity and no CPU or I/O facility was servicing any other activity within the address space.

Warning

Important

High CPU solo time represents a performance improvement opportunity, because it has a direct impact on the run time of your job. Also, depending on the availability of additional CPUs, you might be to restructure your program to take advantage of more than one CPU at one time.

SOLO IN I/O

The percentage of run time spent by the I/O facility in servicing the file access activity, during which no I/O facility or CPU was servicing any other activity within the address space. The I-O-Facility-Utilization-Summary-report further subdivides solo I/O facility usage for each file access activity by unit and volume. For task execution, the I/O solo time is always zero. High I/O solo time represents a performance opportunity. See Report-terminology for more information on interpreting this field.

SOLO IN EITHER

The percentage of run time during which one activity alone was being serviced for the measured job step, whether the service was by the CPU, an I/O facility, or both. SOLO IN EITHER represents the sum of:


    • CPU solo time percentage for the activity
    • I/O solo time percentage for the activity
    • Percentage of run time during which both CPU and I/O facilities were servicing the activity but no other activity was being serviced.

CAUSING CPU WAIT

The percentage of run time during which execution was delayed pending the occurrence of an awaited event, usually an I/O completion event.

For a task execution, CAUSING CPU WAIT represents the time during which execution for the measured job step was blocked, waiting for a completion event not associated with any particular file access. In the case of online subsystems, this time is most commonly simple terminal I/O wait—wait incurred because there are no transactions to process.

Other factors that may contribute to wait time reported in .FILEMGT include tape mounts, module loading, inter-region communication, operator replies, and timer requests. If the module in which the wait occurs is not a standard access method module, the report includes in a line entry for a task time spent waiting for completion of any of several file access operations: a multiple wait condition. Normally, this percentage of wait time cannot be reduced.

For file access activities, CAUSING CPU WAIT represents the time during which execution in the measured job step is blocked while waiting for an I/O operation to complete. Normally, this value is greater than or equal to the SOLO IN I/O value for the activity. It includes the time during which the activity was enqueued because of contention in the system for the required I/O facility. It does not include time spent waiting for data sets to be opened. (Such activity is reported in the .FILEMGT line.)

Among multiple waiting tasks, Strobe assigns wait to the task that is suspended because of file I/O.

Time spent waiting in a standard access method module for the occurrence of one of several events—for example, for the completion of OPEN and CLOSE operations—is assigned to the .FILEMGT file access pseudo-activity.

Warning

Important

If there is excessive contention from other active address spaces for the required I/O facility, the activity may be delayed while the system enqueues I/O requests before the initiation of I/O operations. The report shows the delay in a much higher CAUSING CPU WAIT value than that for SERVICED BY I/O. In this case, implementing IBM-developed I/O avoidance technologies such as caching, LSR pooling, hiperbatch, and BatchPipes may reduce wait time.

CUMULATIVE PERCENTAGE SOLO TIME

The cumulative total for each of the values in the SOLO IN EITHER column, for line entries down to and including itself.

CUMULATIVE PERCENTAGE CAUSING CPU WAIT

The cumulative total for each of the values in the CAUSING CPU WAIT column, for line entries down to and including itself. The last cumulative total equals the WAIT TIME percentage in the Measurement Session Data report.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

BMC AMI Strobe 21.01