Poor change management causes network outages

Insightful analysis by consultants Steve Taylor and Jim Metzler, plus links to the latest WAN news headlines

In a recent newsletter, we expressed our frustration that many of the fundamental problems facing IT organizations today are the same as they were decades ago. For example, applications that were originally developed to run over some form of high-speed, low-latency LAN and later deployed over a relatively low-speed, high-latency WAN don't perform very well. Today, we'll address another problem that just does not seem to go away: the impact of change management on network availability.

To put this subject into context, it would be difficult to overstate the importance of network uptime and performance to the vast majority of organizations. On an ever-increasing basis organizations run their key business processes over their network. As a result, if the network is not available or if it is not performing well, the organization's key business processes are severely impacted.

To ensure high availability, most IT organizations design their network with the goal of minimizing single points of failure on the end-to-end network paths. To accomplish this goal, IT organizations typically deploy redundant configurations of highly resilient network devices complemented by the use of fast fail-over protocols. Minimizing single points of failure clearly increases network availability. This is all standard network design 101.

To continue reading, register here and become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

In a recent newsletter, we expressed our frustration that many of the fundamental problems facing IT organizations today are the same as they were decades ago. For example, applications that were originally developed to run over some form of high-speed, low-latency LAN and later deployed over a relatively low-speed, high-latency WAN don't perform very well. Today, we'll address another problem that just does not seem to go away: the impact of change management on network availability.

To put this subject into context, it would be difficult to overstate the importance of network uptime and performance to the vast majority of organizations. On an ever-increasing basis organizations run their key business processes over their network. As a result, if the network is not available or if it is not performing well, the organization's key business processes are severely impacted.

To ensure high availability, most IT organizations design their network with the goal of minimizing single points of failure on the end-to-end network paths. To accomplish this goal, IT organizations typically deploy redundant configurations of highly resilient network devices complemented by the use of fast fail-over protocols. Minimizing single points of failure clearly increases network availability. This is all standard network design 101.

However, there is a larger issue that impacts network availability that many IT organizations avoid dealing with and Jim addressed that last week when he was in Chicago to moderate two tracks at Network World's IT Roadmap conference. At one of the tracks, "Network Management, Automation and Control," which was attended by roughly 100 people, Jim asked how many in the audience work in an organization in which more than half of the outages are a result of poor change management. Virtually every hand went up. While not surprising, this was discouraging because ineffective change management has been the major cause of network outages for at least the last decade.

In order to truly have a highly available network, IT organizations need to go past what they learned in network design 101. They need to take steps to reduce the human errors that occur when IT organizations make any kind of change to their network. That is a topic that we will explore in future newsletters. It is also a topic that will be discussed in future IT Roadmap conferences and the next one is in Boston on June 18. We hope that many of you can attend.

Read more about lans & wans in Network World's LANs & WANs section.

Steve Taylor is president of Distributed Networking Associates and publisher/editor-in-chief of Webtorials. Jim Metzler is vice president of Ashton, Metzler & Associates.