Extreme was the latest networking vendor to announce their architectural vision for SDN. It is a very open/agnostic approach to SDN, but the most intriguing piece of the announcement (to me at least) was that Extreme’s SDN controller is based on OpenDaylight code. Extreme describes their controller as providing a hardened, enterprise-grade version of ODL, for which they offer commercial maintenance/support.
This certainly adds legitimacy to the ODL initiative. As I blogged in February, ODL needed mainstream networking vendors to “…integrate it into their SDN solutions and/or wrap sheet metal around it and sell it with a support contract.” Well, check that box, as the Extreme announcement fits that bill. In addition, HP recently upgraded their membership status in ODL, and eight others have joined the project since February including Avaya and Oracle. The upgrade of HP’s membership is telling given their open and standards-based approach to SDN.
We’ve been very skeptical at Gartner about the original intentions of ODL, because generally, a bunch of incumbents don’t get together and say “hey, lets create competition and give our customers more choice“. Thus, we were concerned the founders’ original intentions were likely to gain control of and potentially stifle the SDN movement. You can argue that this came true to some degree (a few startups would certainly agree).
However, at this point ODL has become something legitimate and also has graduated from “science project” to something that can be considered for niche use-cases. It is not quite sunshine and daisies yet; ODL still has a long ways to go. The code is in its first (developer) release, and consequently doesn’t have enterprise-grade stability without the backing of someone like an Extreme. This has been confirmed to me from several sources both inside and outside of ODL. Moving forward, perhaps the greatest promise of ODL would be to deliver a standardized northbound API that can be widely used in the industry (I can dream!). Let the comments roll…