WhatsUp Gold: Daily Network Monitor Blog

Network Monitoring News

By David Seuss

Complex networks make it hard to detect issues before service degrades; isolate root causes before veering down the wrong path; or manage a flood of alerts – while not missing the most important ones. These layers of complexity can easily increase mean time to recovery (MTTR) which is the average time that a device will take to recover from any failure, and it’s expensive.

2011-08-29-difficulties

In fact, last year the Aberdeen Group issued a report* that calculated the cost of downtime to average $181,770 an hour.

To fix this problem you may turn to a Big IT Vendor to seek an application performance monitoring (APM) product that promises to resolve issues more quickly in order to avoid downtime and performance issues. In reality the “solution” can become almost as big as the problem – as evaluations take too long, costs run high, features get overloaded, and integration becomes too complex.

Before you get yourself into this situation you might want to consider three success factors that can help you find the best APM product to reduce MTTR:

1. Control
The biggest obstacle to reducing MTTR is finding and isolating the root cause before any compromising event occurs. In order to do this you should be able to define complete dependencies for each application in order to find the quickest path to the root cause.

2. Visibility
One unified dashboard view of conditions across the entire application ecosystem can give you the required visibility into threats against overall health, and the ability to drill straight down to root cause. In order to avoid taking too many steps (and missteps) you should be able to use a single monitor.

3. Agility
Being able to easily find a root cause is a big deal. Once you find it, you have to be able to fix it immediately. In this case, the fastest route to resolution means you should be able to easily set up automated policies and actions.

Be optimistic. You can reduce MTTR by comprehensively monitoring of all of your critical applications, and their dependencies to better detect problems, isolate root causes and restore services.

You just don’t have to break the bank and wait too long for it to happen.

 

*”Datacenter Downtime: How Much Does It Really Cost?” by D. Csaplar, Aberdeen Group, March 2012

 

Share:
  • Facebook
  • Twitter
  • Digg
  • Tumblr
  • StumbleUpon
  • del.icio.us
  • Yahoo! Buzz
  • Google Bookmarks
  • Reddit

Write a Comment