SDNs: The Only Thing to Fear Is Confusion Itself
Technology itself is hard and confusing. But the sophistication of the technology itself doesn’t hold a candle to the complexity surrounding the higher-level issues involved in creating a market for that technology.
Software-defined networks (SDNs) are conceptually easy to understand. By keeping the flowing data separate from the management layer that controls where it all goes, smart real-time decisions based on a wide and inclusive set of variables can be made on how best to route that data.
Sounds simple. But the road from the lab to a network near you is rocky. One problem is that there is no agreed upon approach to SDNs. In addition to this rather glaring issue is that it is impossible to implement a new technique – no matter how beneficial – without exhausting the value of the existing infrastructure. In other words, change takes time because of the investments that already have been made. There is a great deal of money on the table, and how these issues work themselves out will make one vendor a favorite over another. So the competition framed by these variables is fierce.
This is why many engineers don’t like to leave their labs. In that context, results of a recent Brocade survey are interesting. An almost identical percentage of companies currently use SDNs (19 percent) and have absolutely no notion of doing so (20 percent).
The good news for SDN proponents is that the idea is popular for a multitude of reasons. No less than 10 rationales for using SDNs were cited by 15 percent or more of the respondents. The reasons ranged from increased productivity (42 percent) to risk mitigation (15 percent).
While 80 percent of respondents to Brocade’s survey are now or plan to do something with SDNs – and while a tremendous amount of investment is being made – the general perception still is that networking experts are struggling with the concept. Writes The Virtualization Practice’s Steve Beaver:
So why is it taking too long for SDN to get to the point of mainstream adoption? I really think it is following the adoption path of server virtualization in that companies and enterprises are having a little difficulty wrapping their heads around SDN and how adoption of this technology will save on the overall networking costs and at the same time save them time and money with the overall network operations.
He suggests that the key is that vendors haven’t explained their value easily enough:
The big difference that I can see between the adoption of server virtualization and the adoption of network virtualization is that most hypervisor vendors hit the market with all sorts of use cases showing how server virtualization will save organizations time and money in their data centers with smaller footprints; there just does not seem to be the same effort by the network virtualization vendors to present some really compelling use cases.
That probably is true. It also may be true that changing the network plumbing so fundamentally may require higher level sign-in than the much more subtle step of changing how servers are divvied up. That higher level of scrutiny can spell trouble for those trying to make their case. Executives, for the most part, are not known for their technical savvy or desire to take big risks.
Reuters’ Sayantani Ghosh offers a piece with a simple (C-level exec-worthy) explanation of SDNs and what they do. It is a mostly positive piece, but also points to the confusion in the marketplace. The questions going forward are whether that confusion is as bad as it seems from outside IT and telecom departments and how long it will take for the persuasive rationale for implementing SDNs to win out.