Automate performance monitoring
- The EXEC DISA captures IMS status regularly on the Journal log by issuing the commands /DISPLAY ACTIVE and /SSR -DIS THREAD(*) every 15 minutes. The responses are written to the Journal log.
- An EXEC can automatically start and stop DC Monitor for the required monitoring interval.
- A time-initiated EXEC named MIDNITE can reset logs for the next day; for example:
Issue /CHE and /DBR D121PART to force the OLDS to switch.
This action provides a cutoff for daily IRUF processing for IMS PR and IMS TA
- Switch BBI-SS PAS Journal and Image logs
- Reset BMC AMI Ops Monitor for IMS Resource Monitor and BMC AMI Ops Monitor for IMS Online history buckets to zero
- A time-initiated EXEC could coordinate monitors with activity. You could do one of the following tasks:
- Change active monitors when different regions, classes, or workloads are scheduled at various times of the day
- Run a set of special monitors from 11:00 A.M. to 11:30 A.M. with automatic logging to create a peak-usage profile
- Reset monitors at 6:00 P.M. when the BMP workload starts
You can gather workload and service level data from the Image log and write it to a data set for the performance analyst.
This data can include information such as the number of transactions processed and response time, written to the Image log on a timed basis. The batch job can be tailored to select the information needed to solve a specific problem (to be run on request) or to select a standard set of data (to be run on a regular basis).
When the job is complete, a message can be sent to the performance analyst stating that the named data set is now ready for review.
Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*