Performance monitoring - Overview

RMF Performance Monitoring (RMF PM) allows you to monitor the performance of your z/OS host from a workstation through a TCP/IP interface to one or more z/OS sysplexes. You logon to any sysplex and you can monitor the resources in the corresponding sysplex.

RMF PM takes its input data from a single data server on one system in the sysplex, which gathers the data from the RMF Monitor III on each MVS image. Therefore, this is called the Distributed Data Server (DDS). If you want to monitor several sysplexes, each one needs to have an active DDS.

RMF PM provides a selected subset of the information provided by the RMF Monitor III gatherer: general performance data, performance data for jobs, and workload-related performance data like:
  • WLM workloads
  • WLM service classes
  • WLM service class periods
  • WLM report classes
You have the flexibility to create unique scenarios that monitor the performance of your system. You can sample real-time data as bar charts, and combine data from different resources together. Once you have created these scenarios, you can save them as PerfDesks. With PerfDesks, you create a set of DataViews customized to your monitored system(s). DataViews sample performance data into one or more Series displayed as bar charts. You can reuse the DataViews any time. You can simply open the PerfDesk and start it whenever you want to view performance data in your monitored system again from the same angle.
Note: A saved PerfDesk does not contain any performance data. The PerfDesk samples new performance data each time it is opened and started. However, you can save the sampled data in spreadsheet files.

When you open RMF PM, you will find a list of Sysplexes in the resource view on the left side. Each sysplex comprises a hierarchy of Resources of your monitored system(s). Each Resource has a set of Metrics, which specify what is measured, for example, how the Resource is used or loaded with work.

Getting help in RMF PM

All help information will be displayed using the default browser of your installation.

Putting it all together

  • You select Resources and Metrics, add them as Series definitions to a DataView.
  • You save one or more DataViews as a PerfDesk.
  • You start a DataView or PerfDesk to sample performance data on monitored system(s) and display them in the Series of DataView(s).
  • You save the Series to a spreadsheet file.
  • You open PerfDesks any time later to restart performance data sampling.