In my work for IBM on the internal deployment of IBM Lotus Sametime, I’m often asked, “How does IBM monitor its Sametime environment?” Some people expect an answer that reveals a secret, non-intuitive approach to monitoring, and they are surprised by the simple approach IBM takes toward this task. In short, we IBM Sametime administrators use a combination of real-time monitoring and trend analysis of historical data to completely understand our Sametime environment. It doesn’t get any more tricky or complex than that. Yes, we created some collection routines and a display tool to aid our two-pronged monitoring approach; however, the mechanics are straightforward.

In this article, I showcase IBM’s monitoring approach to explain the readily available mechanisms by which we collect both current Sametime usage and server data, and how we make the information available to admins in a unified dashboard plug-in, also known as a miniApp. Admins check this dashboard to see whether trouble is brewing. The environment also delivers historical data for trend analysis in support of assessing infrastructure needs and planning any upgrades or maintenance. I also discuss how we perform additional calculations and then store all data in a Rational database.

To put everything into context, I’ll start with a brief explanation of the overall Sametime server monitoring architecture.