Blog Post

SDN is not OpenFlow, but OpenFlow is a real disruption

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

If 2012 was the year that software-defined networking sold out, then 2013 should be the year that the big players in the industry recognize that their efforts to neutralize the threat of OpenFlow and the coming commoditization of networking hardware are doomed to failure. I’m sure that many people will declare me wrong, but the promise of SDN and the promise of OpenFlow are different.

Software-defined networking doesn’t require OpenFlow. And it will still make a network programmable and responsive in ways that both scaled-out web-services providers and enterprise customers dealing with virtualization will appreciate. But it won’t necessarily affect the underlying networking hardware in the same way OpenFlow can. However, OpenFlow — the protocol that aims to separate the intelligence required to route a packet from the act of moving a packet– can commoditize the switches and routers. And it will have a big impact on the networking vendors such as Cisco, Arista and Juniper.

SDN is good for many things, but not everything.

Nick McKeown, one of the fathers of OpenFlow onstage at Structure.
Nick McKeown, one of the fathers of OpenFlow onstage at Structure.
Right now, that impact has been ignored because many of the benefits of software-defined networking can exist without using OpenFlow. Products from Nicira as well as upcoming offerings from Juniper, Cisco and many other SDN startups don’t depend on the lowest levels of the network. They offer programmability and better ways to manage scaled out and virtualized infrastructure without OpenFlow. And they offer these features in a way that’s far easier for enterprises and even cloud customers to consume. Few businesses have the resources to program and support entirely new networking code for new routers built on commodity hardware, and even fewer want to rip out their existing gear to buy a new OpenFlow-based network.

But it’s early: the promise of real, commodity-based networking gear will not dissipate. Stu Bailey, the CTO of Infoblox (s BLOX) told me last week as his company was launching new software-based networking products that the emphasis on specialized chips and networking hardware is doomed. Both cloud and the enterprise networks are becoming increasingly complicated; not just because of virtualization and scale, but also because of the increasing number of devices at the edge. Smartphones are pressuring networks, but adding sensors and other connected devices as integral business equipment, will require some serious thinking about networking hardware and cost.

“How quickly will a large healthcare org realize that the network is not the collection of routers and switches, but is instead these things connected to the network?” asked Bailey. “And how they interact fundamentally and how they are secured is responsible for the business. With that awareness they need SDN economics and OpenFlow will hasten that.” Hence InfoBlox’s decision to focus on software — software that is OpenFlow compatible.

This revolution won’t come immediately, but it will come.

Others, such as Jim Theodoras, director of technical marketing at ADVA, an optical provider to data centers, have expressed similar thoughts. Theodora has also expressed frustration that so far the message of router and hardware commoditization has been ignored. But for the most part, the existing products on the market and their associated marketing and product managers have been good at loudly shouting down the camp of people who see OpenFlow as a legitimate threat to the hardware makers. And it’s not just marketing. There are many obstacles to deploying an OpenFlow-based networking infrastructure.

Lane Patterson Equinix Kenneth Duda Arista Networks Structure 2012
(L to R) Lane Patterson, CTO, Equinix; Kenneth Duda, Founder, CTO and SVP, Software Engineering, Arista Networks
(c)2012 Pinar Ozger [email protected]
People love pointing out Google’s OpenFlow-based data center communications network as an example of a successful OpenFlow-based implementation, but Google had to develop a lot of its own expertise to make that happen. In fact, Ken Duda, a co-founder and CTO at Arista, accused Google last June at our structure conference of bastardizing OpenFlow to the point where it wasn’t OpenFlow anymore. Duda is from Arista, one of the companies set to feel the pain of any router and switch commoditization, but he’s no marketing parrot.

One obstacle, other than just getting the gear to work, is that the current chips sold by merchant silicon providers aren’t ready to support the most recent and faster versions of OpenFlow. NoviFlow, a startup in Montreal, that just announced an OpenFlow compatible switch that processes information at 100 gigabits per second per second — a significant amount of capacity when we’re talking about the data center market — went with specialty networking processors. It couldn’t build such a high capacity OpenFlow switch with existing chips, so it used network processors from EZChip. It’s still using a specialty chip to make its boxes, much like Cisco (s csco) and Juniper (s jnpr) do.

OpenFlow creates a common networking platform, not lock-in

NoviFlow's 100 Gbps switch.
NoviFlow’s 100 Gbps switch.
However, by using OpenFlow, NoviFlow is anticipating a world when customers can buy a range of switches from different vendors and expect them to work because they are all OpenFlow compliant. And that is when the hardware prices will likely change. Just like x86 processors turned the server market into a battle over new features on a consistent platform, networking gear will soon be about a consistent platform where features matter and vendors can’t lock in their clients.

Marc LeClerc, NoviFlow’s VP of strategy and marketing is anticipating that day, explaining that NoviFlow has a high-end switch, but that it expects customers to also purchase lower-end gear that they will expect will work with the NoviFlow products. And when that day comes, the question is what kind of shakeout we’ll see in the networking world.

“When the world went over from mainframes to client-server it was a huge shift and plenty of companies that used to play in that market like Wang and DEC are no longer around anymore,” Bailey said.

5 Responses to “SDN is not OpenFlow, but OpenFlow is a real disruption”

  1. Have read this a few times now and still like the post, and I think the comments are interesting.

    I think to some extent the theory that commoditization is driving SDN is part of why some of the larger players initially were either slow to adopt or conversely were quick to detract. If you view SDN as an existential threat to your business, it is reasonable to go only begrudgingly (and kicking and screaming).

    But I don’t believe that commoditized networking hardware is driving SDN. With this post in mind, I jotted down some of my own thoughts.

  2. Very well argued! SDN and OpenFlow aren’t perfect in their early stages, but whatever is? The promise of SDN is to abstract those parts of the network infrastructure which need to evolve faster than ASICs in a box and change. Use cases will grow as the R&E Community works on this and startup vendors see opportunities.
    All we are saying, is give these a chance (sorry John Lennon!).
    –Dan O’Hanlon — WVNET

  3. tharvey2013

    While interesting, I think SDN and Openflow still have a ways to go. Thinking from an economic point of view Network vendors still control the industry by restricting barriers to entry.

    The switching costs, network effects, and sunk costs fo building IT networks all contribute to the low adaptation of SDN and Openflow this year

    Let Cloud take greater hold within Enterprises first then you will see SDN and Openflow take off !

    Theo Harvey

  4. Stacey

    Thanks for the interesting article regarding the battle for mindshare in the Software-Defined Networking industry. The necessity of a standardized data plane abstraction protocol, like OpenFlow, is a great subject for immediate discussion.

    There is a small camp of one or more incumbent vendors that have built wonderful 60-70% gross margins, profiting immensely from their customers by maintaing the complexity and obscurity implied with traditionally vertically integrated networking switches and routers. They will fight tooth and nail to maintain that complexity and obscurity that is totally eliminated by a standards-based data plane abstraction protocol, like OpenFlow. They will use every possible marketing resource to spin their vendor-specific extensions of protocols to appear as software-defined networking.

    Meanwhile, there is a growing counter-camp of vendors that is fighting to drive change towards a novel networking architecture that employs standards-based approaches to programming and managing networks via a centralized controller. Such an architecture drives massive operational and capital cost efficiencies for customers by eliminating vendor lock-in to particular vendor hardware. It also allows for dramatic improvements in programmability and automation, which is increasingly needed in the modern cloud data center.

    This is the battle ahead in the quest to control the “Software-Defined Networking” category leadership. As an industry, we can act like sheep and continue to believe the FUD and lies of certain incumbent vendors, or we can be bold and embrace a new architecture that brings networking into the modern world of software automation and programmability. The choice is ours. Let’s beware of the stakes at hand in this emerging battle of architectures.

    Jason Matlof
    VP Marketing, Big Switch Networks

  5. Michael Bushong

    [full disclosure – I work at one of those startup companies looking to leverage SDN in delivering solutions]

    I think the commoditization discussion is slightly off. Commoditization basically means that there are equivalent products whose only differentiation is price. There is nothing inherent in SDN that prevents capability (or even performance once you get into path optimization) differentiation.

    SDN very likely will lead to reduced hardware costs, but those costs don’t disappear. They shift. I would think that the price shifts to orchestration and workflow automation applications built using SDN technologies (OF, PCE, BGP-TE, whatever). What you end up with a shift in pricing mix where the hardware is no longer the dominant monetization vehicle and software is attributed more value.

    This shouldn’t be a surprise though. If you look at the major vendor R&D spends, it is already heavily skewed towards software, even in the vendors that have significant ASIC shops.

    If the networking vendors intercept this by offering more software (and every indication is that they will), then we basically just change the pricing and move on. SDN will give users far more for their dollar, but I don’t think it will lead to a major inflection point in overall solution spend.