
Unexpected increase in response (RESPINCR)
In this scenario, a CICSĀ® systems programmer sees a notification that some of the transactions on one or more CICS systems are experiencing increases in response time.
The reports in this sample might help you identify the cause of the increase in response time.
The report set produces the following reports:
- BADRESP: Response time increase - top 20 worst Response times
This list extended report shows the 20 worst response times for each transaction ID, and how much of the time is Suspend Time and Dispatch Time. - RESPPEAK: Response time increase - Response time peak percentiles
This report shows the average response time of each transaction and the estimated maximum response time of a percentage of tasks. You can use this report to identify any trends of increased response time for particular transactions. - RESPRNGC: Response time increase - Response time distribution
This report shows the number of transactions for each range of response times by interval and transaction ID. Similar to the RESPPEAK Response time peak percentiles report, this report helps you determine if there have been any changes in response time and enables you to identify any specific time periods when the response time was affected. - RESPRNGM: Response time increase - Response time distribution (count & %)
This report is a slight variation of the RESPRNGC response time distribution report. RESPRNGM has additional columns to show the percentage of tasks whose response time is less than 1, and the percentage whose response time is greater than or equal to 1. - RESPRNGP: Response time increase - Response time distribution (%)
This report shows the number of transactions for each range of response times as a percentage of the total number of transactions. Similar to the RESPPEAK report, this report helps you determine if there have been any changes in response time, and enables you to identify any specific time periods when the response time was affected. - RESPWLMP: Response time increase - Response time distribution by Service Class (SRVCLASS)
This report shows you the percentage of transactions that had a response time within specific ranges. The transactions are grouped by service class. - WAIT0001: Response time increase - Wait analysis
The Wait Analysis Report provides a breakdown of wait activity by Transaction ID. You can quickly see which CICS resources are causing your transactions to be suspended. - TRANMNGR: Response time increase - Transaction Manager Global statistics
This report shows if MAXTASK was a cause of the response time increase. If MAXTASK was reached, the report shows the number of tasks affected and how much delay the affected transactions experienced. - DISPOVRV: Response time increase - Dispatcher statistics overview
This report shows how many dispatcher tasks were running in the system at the peak time and the total amount of CPU time the region used. This information can help you determine if a region is underused, overused, or running close to optimum capacity. - TCBMODES: Response time increase - Dispatcher statistics TCB Modes
This report shows TCB event information for each mode, including attaches, detaches, and steals. The report also shows timing information, such as operating system wait time, waits, TCB dispatch time, and CPU times. - TCBPOOLS: Response time increase - Dispatcher statistics TCB Pools
This report shows how many TCBs were attached for each TCB pool. It also shows various wait times that affect the response time of transactions. - MONTORNG: Response time increase - Monitoring Global statistics
This report shows average and peak transaction response times, which provides an indication of how CICS is currently performing. You can use this report to identify CICS regions that might have experienced increased response times.
Parent topic: Problem scenarios and report set samples
