Written by Jim Metzler
Jim Metzler is an expert at researching and analyzing emerging technology trends within the networking and service provider space.
SDN is not flawed, but a fundamental shift to SDN is a slow moving process. Putting things into perspective, it took VMware about 10 years to gain mainstream acceptance. And we’re only in year eight or nine of SDN, so we’re getting close to the tipping point.
A comment I hear a lot is, "I'm tired of SDN". "I'm tired of all this talk, talk, talk. Seminars, white papers, webinars, and it hasn't come to pass." "What's wrong with SDN?" "Is it flawed?"
I think we should put this in perspective. Let's look at virtualized servers, an overnight success. Well, actually, VMware was founded about 1998, and it began to get traction, I'd say, about ten years later. I'd also suggest that stayed inside of one technology domain. Ten years for a fundamental change inside of one technology domain is a worthwhile reference point. I'm not saying that it applies to everything, but it's a nice reference point.
If I look at open flow as kind of a piece of SDN, Martin Cosado got started playing with that, if you will, as a graduate student in Stanford back in 2006, 2007. So I can make a pretty valid argument, we're about year 7, 8, 9 in terms of SDN. We are getting very close to crossing the chasm, so I understand the weariness, I totally do. But it takes a while for a fundamental shift, by definition of fundamental they don't occur quickly. It takes a while to line up all the component pieces and I'd suggest we're still 12, 18, 24 months away from what crosses the chasm. By the way, crossing the chasm does not mean everybody's doing it. It means it moves away from early innovators to more of a mass market. I understand the weariness, but that comes with the territory of fundamental technology change.