Bringing Networking Out of the Stone Age – Insights from the Stanford Open Networking Summit
I recently returned from the Open Networking Summit (ONS), which took place at Stanford University on October 17-19th. It’s clear from the conference that the networking industry is on the verge of a major transition. And leading that change is OpenFlow (or as it has come to be defined in the industry: SDN – Software Defined Networks), which was the star of the summit.
First, a little background; the concept of Openflow was first presented in a PhD paper by Martin Casado at Stanford. Casado looked at the developments over the last 15 years in areas such as the Internet, mobile, applications and so on and realized that these fields are virtually unrecognizable to how they were 15 years ago. He asked why, then, has the networking foundation, on which everything sits, hardly changed at all?
It is stuck in the virtual Stone Age. The implications of OpenFlow for networking companies are tremendous. Currently, companies are occupied with trying to navigate the complexities of the archaic networking system, instead of focusing on developing real innovations (as we see in the world of apps and internet).
Openflow/SDN challenges the basis of the old style networking, and suggests a completely new approach – a centralized algorithm and intelligence, rather than a distributed one (Distributed algorithms are typically executed concurrently with separate parts of the algorithm having limited information about what the other parts of the algorithm are doing). This centralized approach leads to a democratization of networks, which means that anyone who wants to control the network could do so through programming, using an abstraction layer as the network operating system (Network OS). Even if ultimately the specific solution will be different, it is clear that the approach presented by Casado and Stanford IS the future of networking.
Proof of this was obvious at the conference itself as 300 participants (including all the major tech institutions, MIT, Princeton etc.) were in attendance. Plus, more than 500 people wanted to attend, but even the conference’s planners didn’t foresee the tremendous interest. And 20 demos of prototypes were presented on OpenFlow (out of 100 who vied for these spots). At the next meeting of this community (planned to take place in about 6 months), it’s safe to say that the organizers can plan for an even larger attendance.
Radware’s demo raised interest at the conference because unlike the vast majority of the demos, which presented prototypes and betas, we showed an existing, commercially successful product running on the SDN platform, which interfaces with the “Trema” open flow controller. Radware’s DefensePro security product has already been installed on over 1,000 customers’ networks. We also demonstrated Radware’s ADC solutions running on the SDN platform, which drew a large interest as well.
The future clearly emerged last week at ONS, as Openflow will influence every aspect of networking, including application delivery controllers, security, quality of service and more. Today’s networking industry specializes in mastering complexity; it will now transition into creating simplicity. And network professionals, free of the burden of overcoming the challenges of the complex networking, will be free to create true innovations, leading networking out of the Stone Age.