How to know whether a network architecture would be effective in your company

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

In the last couple of newsletters, we've discussed the fact that many companies do not have a network architecture. We also discussed some of the many pitfalls that companies run into when attempting to create a network architecture organization. Today, we will recommend a strategy that companies can use to create an effective network architecture organization.

The key determining factor in whether or not a network architecture is effective is whether the architecture drives decision making in important areas such as the selection of technology and vendors. For the sake of example, assume that a company has a frame relay network in place and intends to stay with that network and their service provider for the foreseeable future. There is relatively little value in creating an architecture for this component of the company's WAN. Put another way, a company will not get a lot of value out of an architecture organization if the focus of that group is to document well-understood, stable technologies that reside largely or entirely within a single IT discipline.

We recommend that a company only create an architecture organization if that organization is focused primarily on providing guidance on emerging issues that cross IT disciplines, could have a sustentative impact on the company, and for which there are multiple approaches.

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 the last couple of newsletters, we've discussed the fact that many companies do not have a network architecture. We also discussed some of the many pitfalls that companies run into when attempting to create a network architecture organization. Today, we will recommend a strategy that companies can use to create an effective network architecture organization.

The key determining factor in whether or not a network architecture is effective is whether the architecture drives decision making in important areas such as the selection of technology and vendors. For the sake of example, assume that a company has a frame relay network in place and intends to stay with that network and their service provider for the foreseeable future. There is relatively little value in creating an architecture for this component of the company's WAN. Put another way, a company will not get a lot of value out of an architecture organization if the focus of that group is to document well-understood, stable technologies that reside largely or entirely within a single IT discipline.

We recommend that a company only create an architecture organization if that organization is focused primarily on providing guidance on emerging issues that cross IT disciplines, could have a sustentative impact on the company, and for which there are multiple approaches.

A great place to start would be by developing an architecture for application delivery. The performance of applications has a very sustentative impact on a company and the topic clearly cuts across all of IT. As we have documented in multiple newsletters, there are many approaches to this problem. Some of the fundamental questions that this architecture would have to address include what do you do yourself and what do you outsource to a third-party? What functionality do you implement at Layer 3, Layer 5 or Layer 7? How do you integrate your application delivery strategy with your security strategy to avoid situations where the data gets encrypted by the application and hence all of the application acceleration appliances that you have deployed no longer function?

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.