Don't Ignore Performance Management Automation

I'm not at all surprised by the results of the mValent survey. Clearly IT is being asked to do more with less labor spend and automation is the only way to achieve this. While configuration changes may take 11 people to make, these are usually discrete tasks with clear handoff, so it is not like 11 people are involved concurrently for 12 or 15 straight hours. However, that is exactly what happens when a mission critical, customer facing application goes down or experiences poor performance. It is not unusual for a dozen or more IT professionals to be on a bridge call for 12-15 straight hours trying to figure out what is wrong and how to fix it. Because each technology silo expert is evaluating alerts from their chosen static threshold-based monitoring solution, the analysis that gets organizations to root cause of these problems is manual, time-consuming and uncollaborative. Sometimes root cause is never found as IT staff desperately re-boot server after server until the problem goes away.

Performance management in today's complex IT environment requires automation through real-time analytics. Real-time analytics can learn the behavior of applications and get IT staff to the root cause of problems without all the manual effort and tribal knowledge-based guesswork. Many organizations are starting to look at these technologies to streamline performance management and to predict outages and brownouts before they affect end users and the bottom line of the business. Every week I read about more and more outages and brownouts that affect the reputations and bottom line of businesses like Amazon, Apple, Twitter and Google. These organizations must take a proactive stance to mission critical service and application performance management and they need to do it NOW! Certainly the consumers of their services would appreciate it... or maybe they'll just go elsewhere...

Click to read the article this is in response to.