SDNs: Big Deal or Big Hype?

Researchers and analysts like to say that software-defined networks (SDNs) are the wave of the future because they separate the data plane from the control plane. What that means, in simpler language, is the data that is being transmitted is independent of the intelligence controlling how it gets from point A to point B and beyond.

That’s easy to understand in theory. But, shocking as this may be, it turns out to be very complex in reality. In fact, two of the most important things for non-engineers to understand about SDNs is that there is no unanimity on precisely what they are and how they will be implemented, and that billions of dollars rest on the nuances of how these issues are settled.

Right now, those subtleties are being worked through and the jockeying in on in earnest. Light Reading’s Jim Hodges — in a story hyping a report from the organization — tackles the question of whether SDNs are game changers or just another interim step in the road ahead. He offers two points suggesting that they are transformative (they are being driven by real-world developments and can simplify complex networks) and that they will fall short of the most extravagant expectations (implementation is difficult and the jury still is out on vendor opinion of SDNs).

Most telecom and IT folks have heard a lot about SDNs during the past few months. That stream of news shows no sign of abating. On July 30, Oracle acquired a company called Xsigo. (Somebody who has been around the tech world for a while may say that deals could be rated on the difficulty of pronouncing the name of the acquired company. If so, Oracle picked a winner.) A week earlier, virtualization company VMware bought Nicira. eWeek covers both deals.

Aside from the prudence of the deals themselves, it is worthwhile to note that one sign of a technology’s acceptance is when the small fries start to be gobbled up by the bigger fish. It means that two questions have been answered, at least for the acquiring company: The technology is worth pursuing and it is better to buy than build expertise.

At the beginning of his commentary at Computer Weekly, Forrester Senior Analyst Andre Kindness equates SDNs with OpenFlow, which is just one version. Whether this is totally appropriate or not isn’t important, however, since Kindness talks in general terms in the remainder of the commentary about the best ways to prepare infrastructure and operations teams for the switchover. He offers five suggestions, which range from standardization of process, procedures, roles and responsibilities to deployment of more monitoring tools.

The bottom line is that the switchover to SDNs will not be insignificant. Kindness sums it all up in this paragraph:

This, in turn, means I&O teams need to coordinate infrastructure elements, such as switches, firewalls, load balancers and optimisers, to deliver the right set of services to the right user, at the right time and at the right location. Workload centric networks will reconfigure these elements on the fly and monitor the output to ensure that the newly created services are within the bounds of the business policies and rules.

Kindness points out pretty significant issues. It circles back to the subject of Hodges’ commentary: There truly will be big changes if the industry adopts SDNs en masse. However, there have been many technologies that were thought ready to significantly reorder the landscape — and never did.