Good news about route management

Readers of this column might recall my repeated homilies – OK, you can call them rants – about the need for reducing operational costs in large networks.

Time for some good news: A new product category of “route managers” is poised to have an effect on the cost of operating large routed networks. Assuming the technology works as advertised, beneficiaries include service providers and large companies, which can reduce the amount of time required to configure, troubleshoot and manage IP networks. Oh, and these products will help make IP services more reliable and resilient. Can’t get much better than that!

In a nutshell, the idea is that you drop a box (or boxes) into your network. By listening to the ongoing routing “conversations” in the network, these boxes quickly create a Layer 3 map of the network showing route locations. As the routes change, so does the image – letting you keep a real-time view of the network.

These products aim to alert you early to problems such as an increase in route flapping, which often presages a network outage. They provide real-time and retroactive root-cause analysis capabilities, letting users uncover and address the sources of current and past problems. They aid with scenario-impact analysis, letting you do what-ifs and capacity planning before making changes in your network. In short, they provide network engineers with something they haven’t had before: the ability to visualize graphically what’s going on in their networks.

Note that I’m not talking about router management, which generally focuses at looking at such things as queue depths and port availability. Nor am I referring to element management systems such as HP OpenView and others, which focus on tracking the health and availability of diverse network resources. Finally, these devices shouldn’t be confused with so-called route optimizers, which pass traffic along different routes according to predetermined route selection criteria.

The breakthrough is that the devices provide, in essence, a router’s-eye view of an IP network, both in real time and from a historical perspective. Oddly enough, such functionality has been missing since the dawn of the routing age. Even in the most sophisticated network operation centers at the largest companies, up-to-date route information all too often exists only in the minds of engineers or scribbled onto barely legible whiteboards.

All this translates into two things: cost savings and increased customer satisfaction. By making networks easier to troubleshoot, route managers save valuable engineering time. By making problems easier to avoid and recover from, they improve service-delivery metrics such as mean-time-to-repair, latency and uptime.

The catch? This technology is currently available only from a couple of start-ups – Packet Design and Ipsum Networks – so network executives might face a greater-than-usual challenge in introducing the technology into their networks. Packet Design is exhibiting this week at NetWorld+Interop in Las Vegas. Ipsum received a “new product achievement award” award in January at Comnet.

Johnson is president and chief research officer at Nemertes Research, an independent technology research firm. She can be reached at johna@nemertes.com.