EMA Blog Community

Traditional APM products require significant care and feeding. Customers are required to tune thresholds, for example, to notify when desired performance thresholds are breached.

In contrast, today’s leading edge APM solutions are starting to incorporate self-learning capabilities that enable them to “watch” the IT environment and “learn” normal performance levels for specific times of day, days of the week, and times of the month, for example. These capabilities not only automate the process of setting thresholds, they provide much more fine grained monitoring capabilities than manual configuration can. Often, such products detect environmental characteristics that IT administrators are not even aware of.

In an earlier blog entry, I discussed the idea of analytics increasingly becoming a competitive differentiator for enterprise management vendors.  Multiple APM vendors– CA Wily, Compuware, AppDynamics, BMC ProactiveNet, Prelert, Netuitive, OpTier, IBM Tivoli, Integrien, NetScout, DynaTrace, and Precise – are embedding self-learning capabilities in their products already. It is likely that such capabilities will continue to be added and enhanced throughout 2011.

Self-learning capabilities mean less manual work on the part of IT specialists maintaining the management software. They also mean the APM process becomes more automated and far more precise.

I have been writing about the idea of autonomic computing for five years now, and continue to watch vendors and products to evaluate how the industry is progressing in that regard. In addition to advancements in self-learning capabilities, 2010 also saw the growth of Complex Event Processing (CEP). We also began to see overlay products that gather and correlate metrics from multiple heterogeneous technology sources into a cohesive view of the software enterprise. It appears that most of the underlying elements are now in place to make autonomic computing a reality. The task of the next few years will be to pull the pieces together as viable and marketable products.