Infoblox CTO launches free tool for measuring network complexity

This article was originally published on Technology.Info.
As part of our continuing strategy for growth, ITProPortal has joined forces with Technology.Info to help us bring you the very best coverage we possibly can.

Network complexity, not bandwidth, is the real barrier to scalability in many IT infrastructures, according to Stu Bailey, chief technology officer at network control specialist Infoblox, and it’s a problem he’s determined to help IT teams solve.

For many, it’s a matter of urgency, he says: “When you hear horror stories about network downtime, you can bet that network complexity is the problem. I see a growing awareness that, if network complexity is a problem today, then the adoption of trends such as virtualisation, cloud, mobile and Internet of Things is only going to make things worse, by introducing new traffic and new endpoints to already-complex network infrastructures.”

With that in mind, Bailey has got together with Professor Robert Grossman, a big data expert and academic at the University of Chicago (and Bailey’s old boss and long-time research collaborator), to develop Tapestry, an open-source tool for measuring network complexity. This is not a commercial Infoblox product, but instead, is expected to be available in October as a free download through the FlowForwarding.org online community.

Tapestry is based on a formula for network complexity predicated on the number of endpoints on a network and how they interact to perform key business functions, based on positive DNS [dynamic network services] responses, rather than the more traditional approach of counting the number of infrastructure elements - such as switches and routers - and mapping the wires that connect them. This information will help guide IT teams’ planning and budgeting activities, says Bailey.

It’s basically a software-defined network (SDN) application that uses inexpensive ‘white boxes’ - inexpensive commodity servers - with programmable Ethernet processing from high-volume chip vendors such as Intel and ARM to collect the data. By monitoring changes over time, as measured by the Network Complexity Index (NCI) formulated by Bailey and Grossman, IT teams can better understand their network’s “complexity growth curve”, which will help to guide their network planning and budgeting activities.

It’s early days for Tapestry: in order to measure complexity, it’s been necessary to “start simply”, says Bailey, trialling the tools with Infoblox customers and on the company’s own network infrastructure. It’s too early, he says, to reveal any results from these trials, let alone disclose an average NCI score among these companies.

“We’re not declaring victory just yet. This is just the beginning of a conversation that we hope the whole networking industry will join us in,” says Bailey.

But another benefit he hopes that Tapestry will deliver to companies that deploy it is the chance to get hands-on experience in moving from hardware-defined networking to SDN. New, low-cost SDN devices can instantly transform themselves into whatever the network needs, he says, be that a new router, a new switch, a new firewall or a load balancer. Being able to quantify network complexity for the first time with measurements such as NCI, he says, is an early example of how SDN applications may help CIOs cope with the increasing pace of business processes and the move to virtualised, cloud and software-defined infrastructure.

Topics