Planning your monitoring schedule

You should plan for the following broad levels of monitoring activity:

Dynamic monitoring

Dynamic monitoring, is "on-the-spot" monitoring that you can, and should, carry out at all times. This type of monitoring generally includes the following:

Daily monitoring

The overall objective here is to measure and record key system parameters daily. The daily monitoring data usually consists of counts of events and gross level timings. In some cases, the timings are averaged for the entire CICS system.

Weekly monitoring

Here, the objective is to periodically collect detailed statistics on the operation of your system for comparison with your system-oriented objectives and workload profiles.

Monthly monitoring

Monitoring for the future

When performance is acceptable, you should establish procedures to monitor system performance measurements and anticipate performance constraints before they become response-time problems. Exception-reporting procedures are a key to an effective monitoring approach.

In a complex production system there is usually too much performance data for it to be comprehensively reviewed every day. Key components of performance degradation can be identified with experience, and those components are the ones to monitor most closely. You should identify trends of usage and other factors (such as batch schedules) to aid in this process.

Consistency of monitoring is also important. Just because performance is good for six months after a system is tuned is no guarantee that it will be good in the seventh month.

Related tasks
Performance monitoring and review
Deciding on monitoring activities and techniques
Developing monitoring activities and techniques
Planning the performance review process
Reviewing performance data
Typical performance review questions
Confirming that the system-oriented objectives are reasonable
Anticipating and monitoring system changes and growth
[[ Contents Previous Page | Next Page Index ]]