Real World to SDN Vendors: Give Us a Break, Already

Steven Hill
Steven Hill

Summary Bullets:

  • There’s no question that SDN can solve a number of problems facing the next generation of network administrators, but those problems may not be as earth-shaking as vendors would like you to believe.
  • SDN solutions, regardless of vendor, have been facing slow adoption, perhaps because the pain points they address may only be minor irritations to the typical enterprise today.

Those of us who have been involved in the IT industry for more than a decade or two have seen some pretty substantial changes to the fundamental way we get things done. Major tectonic shifts such as graphical user interfaces and server virtualization have reshaped the way we build our infrastructure by abstracting all the underlying minutia that it actually takes to get something done. As a rule, however, those types of changes impact a large set of technology users on a much grander scale by efficiencies that are easily measureable. Yet, as far as SDN goes, the operation of the network itself is the responsibility of a very small group of individuals who are charged with keeping the system up and running to whatever number of ‘nines’ your business requires, so whatever benefits SDN may actually provide will only directly affect them… with the hope that the rest of the benefits will appear in the form of improved network performance that may or may not be apparent to the end user.

Most SDN solutions depend on adopting a new and relatively untested network environment – or at the very minimum a difficult-to-justify investment in new technology – something from which any network admin in his right mind would probably shy away. The primary goal for administrators is to keep things up and running, and the major rule of thumb is: “If it ain’t broke, don’t fix it.” So, there it is: a major technology shift that focuses on one of the tiniest groups in the IT community, offers a questionable risk-to-reward ratio, and addresses issues that may not really be much of a problem in the typical network environment. I myself would probably hold back in that scenario too, but sadly, I have already heard from some SDN players who think that the hesitance to accept the ‘wonders’ of SDN is mostly about network admins protecting their jobs – a specious argument that is not fair to the people who are responsible for keeping things running 24/7/365.

It occurs to me that the current hype over SDN may be a remarkable case study of the tail wagging the dog, and the real question is: “Who or what is driving the tail?” A few weeks ago, as I was preparing to go on stage to moderate a panel on the future of SDN, I was stopped by an attendee who asked me, “If SDN is supposed to be a user-driven movement, just who are the users you’ve been talking to? Because we really don’t have much of a problem with most of this stuff at all right now.” Yikes. Now, that is the kind of feedback that wakes you up like a cold shower, and I really did not have a good answer. Apparently, no one does, yet every vendor in the industry is scrambling to present a cohesive SDN strategy, in part because the absence of one is starting to become some sort of death warrant. I do not completely agree with all the buzz about the importance of SDN, but I do know that I have yet to hear from a network admin who lies awake at night fretting about his lack of an SDN solution when it is far more common to fret about getting that dreaded, 4:00 AM support call – a delight that SDN does not particularly promise to eradicate. For SDN to start making real inroads in the enterprise, network vendors need to find a way to articulate the value proposition of SDN far more clearly for environments where the benefits may not be readily apparent and to the people who keep things running.

One thought on “Real World to SDN Vendors: Give Us a Break, Already

  1. Hi Steven,

    A well written and argued post. However the need for SDN, and the business case for it, depends a lot on your perspective.

    Network guys need to get out of the mindset, that if it ain’t broke, don’t fix it. From the business perspective, just because your network is up 24×7, it doesn’t mean it “ain’t broke”.

    Let’s say I’m a Business Application owner at a bank, and I want to launch a new sexy type of SmartPhone app for my high net worth clients, that will give me a competitive advantage for 3-6 months or more.

    The virtualised Server and Storage environments can give me what I need in 10 minutes. But it can take 2 weeks for the network and security teams to provision the firewalls, load balancers & network services. Because it’s all still manual. Because it’s all still Physical. Your network may be “up”, but it’s still broken, from my perspective.

    Yes, virtualisation saves Capex and better yet, Opex. But the big bang here for the business is Agility. And that is why we want the entire Data Centre to be virtualised.

What do you think?

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.