Establishing monitoring procedures

Several types of monitoring strategies are available.

You can:

• Summarize actual workload for the entire online execution. This can include both continuous and periodic tracking. You can track total workload or selected representative transactions.
• Take sample snapshots at peak loads and under normal conditions. It is always useful to monitor the peak periods for two reasons:
  • Bottlenecks and response time problems are more pronounced at peak volumes.
  • The current peak load is a good indicator of what the future average will be like.
• Monitor critical transactions or programs that have documented performance criteria.
• Use the z/OS® Workload Manager to help manage workload distribution, balance workloads, and distribute resources.
• Use IBM® z/OS Workload Interaction Correlator to gain unified workload performance data for IMS systems and other participating z/OS components and middleware across the z/OS stack.

Plan your monitoring procedures in advance. A procedure should explain the tools to be used, the analysis techniques to be used, the operational extent of those activities, and how often they are to be performed.

Regardless of which strategy you use, you need to:

• Develop performance criteria
• Develop a master plan for monitoring, data gathering, and analysis