Using Behavioral Analysis and Predictive Analytics to keep SLA’s on Target, Prevent (Recurring) Service Problems, and Reduce Costs in a Mainframe Production Environment

In this presentation I will explain some of the basics of behavioral analysis and predictive analytics describing how these techniques are used for performance management in a mainframe environment. Using data from standard mainframe monitors (e.g. OMEGAMON, TMON, Mainview, SYSVIEW) these techniques can be applied in a standard mainframe production environment to discover performance anomalies. Analyzing these anomalies increases service levels and SLA adherence by finding performances problems earlier (before they affect users). They also lower costs and speed problem resolution by providing the information needed to quickly pinpoint the right experts to fix the problem. I will also briefly touch on how these techniques can be used to manage the costs associated usage based pricing (e.g. sub-capacity pricing).