Where are we with autonomic computing?

* Automating management processes

Is autonomic computing an idea whose time has come? Or is it still in the realm of fantasy? EMA is in the process of answering these questions with multi-year research projects. Our 2006 research into end-user perceptions and acceptance of autonomic capabilities is hot off the presses. And later this year, we will publish a follow-up study detailing the capabilities built into enterprise management products currently in the marketplace.

You may remember that the term "autonomic computing" comes from the body’s autonomic nervous system. This system helps the body to adapt to its changing environment by regulating involuntary responses like heartbeat, blood pressure and breathing. Autonomic computing incorporates a similar idea. Autonomic systems process and correlate real-time environmental data arriving from the infrastructure, then take appropriate action to self-regulate.

The term originated within IBM when, in 2001, an executive proposed the concept as a way to mitigate the impact of complexity on technological growth. As a result, the vendor community tends to see the phrase as an "IBM term", with HP substituting “Adaptive Enterprise” and Microsoft coining “Dynamic Systems Initiative”. Regardless of the name, many products in the marketplace today incorporate autonomic capabilities to some degree.

While I did two webcasts with IBM on this subject (replays are here under the Events link) during 2006, as far as I know the user acceptance survey we just completed was the first of its kind in the industry. More than 150 IT professionals responded, helping us to examine end-user attitudes towards autonomic concepts. The resulting paper, entitled "Is the Industry Ready for Autonomic Computing?" has just been published and I'll be doing a webinar detailing our findings later in January (the registration and replay link is here).

Autonomic computing is definitely a continuum, with our industry currently positioned at the far left of the curve. However, the subject is timely because of the rising cost of IT support. Throughout 2006, this was fodder for frequent comment, with an estimated 60% to 80% of average budget going towards administration and support. Although our research shows that both executive and technical IT professionals are concerned about this number, nobody is surprised. They are supporting increasingly massive, diverse, distributed and integrated technology systems, and this complexity is reflected in bottom line cost.

Since complexity isn't likely to decrease, turning to technology to help solve the problem makes good sense. Our research shows both executive and technical professionals agreeing with this idea in principle, although technical personnel are more wary of actually putting it into practice. It also shows that vendors can mitigate this distrust by building in visibility to autonomic activities. Audit trails, consoles and policy-based configuration are examples of functionality that users want to see before actually considering a purchase.

In general, it looks as if many of those surveyed are ready for a change. Shifting routine, repetitive tasks from people to technology is an idea whose time has come. However, for these products to be widely accepted, they have to incorporate functionality that makes them palatable to those responsible for keeping today's complex execution environments up and running.


Copyright © 2007 IDG Communications, Inc.

The 10 most powerful companies in enterprise networking 2022