MQ Series Activity

The MQSeries Activity report provides a summary of activity caused by application modules and activity by queue. Click on either tab to view the reports.

Activity By Module

The MQSeries Activity by Module report shows MQSeries activity caused by the application modules during a Strobe measurement session as well as the wait time due to making MQSeries service calls.

  • If you set the MQTRAN reporter parameter, this report will contain data showing IMS transaction level activity.

  • If you set the threshold value in the MQSERIES CALLS field of the Strobe TAILOR REPORTS panel, the Total value may not match the sum of the values reported for each call. Strobe does not report detail lines for calls that show less wait time or CPU  than the threshold value that you specified.

Report details

  • Queue Manager: Name of the queue manager of the queue that was the target of the MQSeries service calls.

  • CPU % Total: MQSeries activity with or without any such concurrent activity.

  • Wait % : All causes of wait time, including page retrieval, programmed I/O operations, and timer requests.

Click next to a Queue Manager to view a list of queues that were accessed via this Queue Manager during the measurement.

  • Module: The name of the module making the MQSeries service call. If you specified the name of the common user module on the Strobe - DATA COLLECTORS panel, it will not be listed in this report.

  • Section: The name of the control section making the MQSeries service call.

Click next to a Module to view additional detail:

  • Request Location: Storage location of individual service call in the control section.

  • MQSeries Call: The type of service call made by the control section.

Click next to a Request Location to view additional detail:

  • Queue name: Name of the MQSeries queue holding the data that was the object of the service call. In most cases, this name is the name of the local queue. For a MQPUT1 call, however, this name is either the local queue name, or the alias name (if an alias was used for the queue).

Note: The message 'UNABLE TO IDENTIFY QUEUE' may appear in the Queue field if certain control blocks are not available to Strobe. MQSeries control blocks may not be available at OPEN and CLOSE time. The message 'NO QUEUE APPLICABLE FOR THIS API' may appear in the queue name field when Strobe is processing either internal APIs or MQCMIT and MQBACK APIs.

Activity By Queue

For each MQSeries queue manager, the MQSeries Activity by Queue report provides a summary of the messages received on a MQSeries queue. The report shows the activity caused by application modules that access MQSeries and the wait time experienced by application modules that make service calls to an MQSeries queue. It also provides information about the characteristics of the messages associated with the application activity.

Report details

  • Queue Manager: Name of the queue manager of the queue that was the target of the MQSeries service calls.

  • If the queue was opened using an alias, the alias name displays next to the queue name.

  • If the queue was opened using a model queue, the name of the model queue displays next to the queue name.

  • Otherwise, the queue is assumed to have been accessed directly by the application and only the queue name displays.

  • MQI Count: The total number of MQI Calls seen during a measurement session.

  • CPU % Total: MQSeries activity with or without any such concurrent I/O activity.

  • Wait % : All causes of wait time, including page retrieval, programmed I/O operations, and timer requests.

Click next to a queue manager to view queues within that queue manager. At each observation the current depth for the queue is recorded. The Queue Depth statistics are the summary of the observed events as follows:

  • Queue: Name of the targeted queue.

  • Queue Depth Average is the total of the observed queue depths divided by the number of observations where the queue depth was recorded.

  • Queue Depth Median is the middle of the order set from low to high of the observed queue depths where queue depth was recorded.

  • Queue Depth Maximum is the highest queue depth found during the observations where the queue depth was recorded.

  • MQI Count: The total number of MQI Calls seen during a measurement session.

  • CPU % Total: MQSeries activity with or without any such concurrent I/O activity.

  • Wait % : All causes of wait time, including page retrieval, programmed I/O operations, and timer requests.

Click next to each Queue to view characteristics for that queue. Note that the detail fields just report information about messages that were observed by Strobe. A message not observed could have one of the noted characteristics, but was not sampled by Strobe.

  • Module: Name of the module making the MQSeries service call.

  • Section: Name of the section making the MQSeries service call.

  • Priority Range identifies the range of priorities for the MQSeries messages observed by Strobe for WebSphere during measurement.

  • Message Size:

    • Smallest and Largest: the size of the smallest and largest messages sent to or received by the module section. If the value is 1- 999,999 bytes, a 'B' appears after the number. If the value is one megabyte or more, an 'M' appears after the number. In the case of MQPUT and MQPUT1 calls, this value is the size of the message specified in the service call. For MQGET calls, this value indicates the buffer length specified in the service call.

    • Average: the average size of the messages sent to or received by the module section.

  • CPU% Total: MQSeries activity with or without any such concurrent I/O activity.

  • Wait %:

  • Page: Wait time that results from retrieving a page from the page data set. A high value in this field indicates that there is not enough physical memory assigned to the address space. If you noticed a high paging rate on the Measurement Session Data report, the Wait Time by Module report enables you to see which module was experiencing delay because of paging.

  • Total: All causes of wait time, including page retrieval, programmed I/O operations, and timer requests, expressed as a percent.

  • MQSeries Call: Name of the WebSphere MQ service call. The number of times that the call was made is reported for the following types of calls:

    • MQOPEN

    • MQCLOSE

    • MQPUT

    • MQPUT1

    • MQGET

    • MQINQ

    • MQSET

  • COA: whether a message that was written specified that a confirmation of arrival (COA) should be generated when the message has been put on its final destination queue.

  • COD: whether a message that was written specified that a confirmation of a delivery (COD) should be generated when the message has been retrieved from its final destination queue.

  • Browse: On a MQGET call, the message was copied and possibly locked, but not removed, from the queue.  Valid values include Y or L.

  • Persist: Indicates whether the message is persistent or not.  A message with persistence is maintained on backup storage and remains there across a system failure or restart event.  A message without persistence disappears in a failure or restart event.  Valid values for this column are either a Y or N.

  • GroupId: Indicates whether or not the message was seen as part of a group. Logical messages having the same group identifier are considered part of the same group. Valid values are Y or N. If the value is Y, then the message seen was part of a group.  If the value is N, then the message seen was not part of a group.

  • Data Conv: One or more MQGET service calls specified that the application data in the message should be converted to conform to the values in the message descriptor.

  • Wait: An MQGET call was issued with a specification of a maximum time that MQSeries is to wait for the specified message to arrive on the queue.

  • Corl ID: One or more service calls specified a correlation ID.

  • Sync Point: One or more messages were read or written with a synchronization point specified.

  • MQI Count: The number of times a WebSphere MQI call was made.

  • CPU% Total: MQSeries activity with or without any such concurrent I/O activity.

  • Wait %:

  • Page: Wait time that results from retrieving a page from the page data set. A high value in this field indicates that there is not enough physical memory assigned to the address space. If you noticed a high paging rate on the Measurement Session Data report, the Wait Time by Module report enables you to see which module was experiencing delay because of paging.

  • Total: All causes of wait time, including page retrieval, programmed I/O operations, and timer requests, expressed as a percent.