|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Editions | myCNN | Video | Audio | Headline News Brief | Feedback | ![]() |
![]() |
|||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Taking event correlation seriously
(IDG) -- There has been a lot of attention placed upon event correlation in the last year - by the industry, by some users and, admittedly, by this column. A string of recent news over the last few months has heightened this attention: Visual Networks just announced plans to buy Avesta; Micromuse announced plans to acquire Calvin Alexander; Tavve introduced e-probes; and RiverSoft introduced the concept of a "network management operating system." Event correlation reduces alarms and identifies points of failure across networks, systems and (more rarely) applications. Because these functions are so useful, the recent attention shouldn't come as too much of a shock.
Between 60% and 90% of the time IT managers spend resolving problems is lost to diagnostics. Event correlation promises to significantly reduce that percentage - bringing down operational costs for IT, and reducing revenue lost to downtime by many millions of dollars for large businesses. Also, a correlated knowledge base can improve accuracy and efficiency for other applications, such as those targeted at trending, performance and service-level management. In the most intricate of examples, information from a correlated, self-adaptive knowledge base can inform such actions as software distribution, configuration and change management - pushing towards pretty much the whole range of management disciplines. On the other hand, most users still don't have great confidence that event correlation - and its more mystical twin, "root cause analysis" - actually work. Rules-based approaches - in which armies of well-meaning specialists camp out to map IT eccentricities to a set of raw abstractions - haven't really caught on. More automated approaches have their apostles in the market, but many users still remain restless, skeptical and unsatisfied. No one, not even the marketing wing of the most optimistic event-correlation vendor, claims total victory - so they're not 100% disconnected from enterprise network managers. But they're close. "Automated" solutions often just aren't that automated. Pinpoint accuracy can turn out to be more like barn-door target shooting. Costs are another factor. If a platform costs $15,000, and a third-party event-correlation solution that inhabits it costs $150,000 - which is the more strategic investment? This time, we are seeking your ideas. Are you among the skeptical and the disaffected? Or are you an enthusiast? Why? The survey can be reached online by going to the Enterprise Management Associates home page. The survey should take about five minutes - 14 multiple-choice questions. It will run for six weeks. After that, of course, I'll write a newsletter looking at the results. If you're frustrated, it's a chance to strike back. If you're a fan, it's a chance to vote. I also invite your comments via e-mail. RELATED IDG.net STORIES: Why it may be time to invest in event correlation (if you haven't already) RELATED SITES: Enterprise Management Associate's Web site | |||||||||||||||||||||||||||||||||||||||||||||||||||||
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Back to the top |
© 2001 Cable News Network. All Rights Reserved. Terms under which this service is provided to you. Read our privacy guidelines. |