IBM Cluster Monitor

1. How can I monitor my cluster performance using IBM Cluster Monitor?

IBM® Cluster Monitor provides a automated framework for monitoring cluster performance. At the back end it collects a predefined set of statistics, which are consolidated from time to time. It has a Web-based component that allows you define desired views and graphs for the collected data.

Back to top

2. How are graphs and views linked to each other?

A view is analogous to a screen of the front end. One or more graphs can be added to the view, and you can define more than one such view. Only one view can be visible in one browser window; multiple views can be displayed in multiple browser windows.

Back to top

3. Can I define my own methods for collecting my own statistics and viewing them in the Cluster Monitor front end?

Yes, Cluster Monitor provides interfaces for third-party collection programs, which work as plug-ins to Cluster Monitor at the back end.

Back to top

4. Need I write the plug-in as a shell, as shown in the examples?

Although the example programs are shell programs, plug-ins can be writen in any language as long as they print the required output.

Back to top

5. What if I cannot get the required ganglia and rrd RPM for my platform?

They are open-source products. You can download the source code and build them for your platform.

Back to top