Sure, it's important to resolve outages fast, and avoid them if possible. But it would be a mistake to ignore the other 99.9 percent of the time when everything is "up" but not necessarily performing as well as it could be. Between the two extremes of everything up and running perfectly and a complete outage, there are infinite shades of gray that color the performance of end user applications running over the network.

Fault management is necessary, but not sufficient. The Check Engine light on your car's dashboard is an important indicator, when it's on, but it communicates an exception condition. You need to monitor the metrics - your speedometer, tachometer and fuel gauges - that tell you how well the car is doing at transporting you from one place to another.

If outages are a rare occurrence, a concentration on device availability is placing too much emphasis in the wrong place. Availability management, in isolation, has already taken the enterprise as far as it can.

Denise Dubie of Network World recently wrote about a group of vendors gathered at Interop in a story entitled, "Is Network Management Irrelevant?"

"I think in general we are on the completely wrong trajectory in management," said Shmuel Klinger, vice president of architecture and applied research in the CTO office at EMC, which acquired network-management vendor SMARTS in 2004. "Things are more complex, there are more moving parts and management as an industry are chasing the wrong trends."

There's more to network and application infrastructure than just being "up" or "down." Networks should be managed using a performance-oriented, top-down approach - a philosophy of "performance first" - that puts the focus back on the end-user's experience.

Here's Dubie again at the Interop panel:

"All agreed network management is attempting to evolve fully from an element-based discipline to an effort to manage across IT silos, such as servers, storage, security, databases, applications and network components, but didn't seem to think the methods used today would work until the culture of IT changes inside enterprise companies." [Emphasis added.]

We've started this blog to help accelerate the needed cultural change within the enterprise. We'll be talking about the idea of performance first approaches to network management quite often in the future.

Want to learn more? Listen to our podcast interview with Jim Metzler, a respected industry analyst with Ashton, Metzler & Associates, and author of a new multi-client study on "Defining an Application Delivery Framework."

Fire up your favorite MP3 player and learn about:

• The role of the networking organization in ensuring application performance


• Modeling the application delivery framework


• The relationship between components of the applications delivery framework


• And more!