1 Comment

Occupy San Jose: OpenFlow and Open Networking

 By: Peter Christy Posted: Sep. 28, 2012

This post is also available in: Japanese

Occupy San Jose - OpenFlow and Open Networking

The modern OpenFlow and Open Networking activities began a decade ago with Martin Casado’s work within the intelligence community implementing improved network security based on flow processing. When Martin went to Stanford for graduate studies he joined with Nick McKeown (his Stanford advisor) and Scott Shenker from UCB. All three were founders of Nicira.

New technology ventures are rarely just about the technology. McKeown, Shenker and Casado always had the bigger aspiration of disrupting the network industry, in part to bring the prices of network hardware way down but perhaps in part just to have a little fun at the expense of the network giant, Cisco.

A decade later it’s interesting and informative to assess the impact. Today it’s fair to say that OF and SDN products haven’t stolen much of Cisco’s revenue. It’s turned out that implementing real networks in OpenFlow is much more complicated than it seemed too many at first, and so far, there haven’t been any “killer apps” emerge to drive this kind of transition within the enterprise network market. So you might conclude that OpenFlow is yet another example of an academic idea that turned out to be less than had been hoped for.

But viewed another way, the OpenFlow effort has been wildly successful at disrupting the industry.  OpenFlow created a lot of VC interest, and created startups that poached some of Cisco’s best talent, or so it seems. Then VMware’s recent acquisition of Nicira gave other “OpenFlow” ventures a big boost and did nothing good for Cisco’s market value, not to mention making our three amigos a lot of money.  But perhaps most interesting of all is the Open Networking Foundation, which at recent count had 77 companies (including Cisco and the other network equipment incumbents) and hundreds of top networking people from those companies putting effort into ONF, all basically in support  of the effort to disrupt the industry. Amazing! That’s sort of like one of the major PAC’s putting their money behind campaign finance reform! Theater and sociology play a major role in politics and revolution and from this perspective it’s literally hard to imagine how OpenFlow and Open Networking could have been more successful. The technology may still need a lot of development to be commercially viable but that doesn’t seem to have diminished the enthusiasm for it or the collective enthusiasm for tipping the network industry.

Power to the People!

Guest Blogger Disclaimer:

The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of SDNCentral.com and SDNCentral, LLC. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.

Checkout more from our Guest Bloggers:

Like our Content? Join the SDNCentral Mailing List

  • This field is for validation purposes and should be left unchanged.

About the Author

.

Peter Christy

Peter is the Research Director of 451 Research’s Networking Practice. For more than 30 years, Peter… More

Connect with the AuthorWebsiteGoogle+

  1. Art Fewell     says:
    September 28, 2012 at 10:40 am

    Great post Peter and I love your perspective … but, I would like to offer an alternative explanation regarding this point:
    “It’s turned out that implementing real networks in OpenFlow is much more complicated than it seemed too many at first, and so far, there haven’t been any “killer apps” emerge to drive this kind of transition within the enterprise network market.”

    I do not think there is any technical problem preventing the killer app, just a level of ineptitude by Cisco competitors that I find Colossal. And I say by Cisco competitors because, one would expect that the incumbent tries to play the slow game. Turning a disruptive innovation into a linear series of innovations may not be forward looking as I would like, but there is no proven model that gives incumbents quantitative assurance they can manage a industry disruption and still remain on top. So though I dont love Cisco’s SDN story, they are doing about what I would have expected.

    But, pretty much all Cisco competitors have completely botched the opportunity. All of the complaints about the flaws of openflow would be completely irrelevant if they would have just done it in the branch first and then expanded as the technology matured, just like lwapp, just like voip, just like most new architectural models. If 2 years ago people would have started showing up in the campus with a nice controller just like what trapeze had in their early days, it would have been just as easy to sell as the first wlan controllers. And I was reselling Cisco FAT AP’s on the ground level back then, nobody was whining about the maturity of the technology because they didnt 1st decide to position it in the middle of the most demanding possible use cases. Selling wlan controllers back then was like taking candy from a baby,m the sales motion was that you showed up, put the cool gui in front of the customer, then they bought it. The drastic nature of the change was exactly the same for wlan as it is for openflow but becasue of the way that the industry executed for wlan, nobody ever feared thin AP’s, they were seen as a simpler alternative from the get-go.

    But the powers that be were so focused on Data Center, they couldnt see a blue ocean opportunity if they were drowning in it. There is NOTHING technical that prevented that from happening, purely politics. Purely because in our hype-driven speculative bubble market 99% of what any big business is doing uses an entirely different risk-tolerance model than even the most conservative VC’s. Granted they all have their little VC experiments, but within every market one question remains supreme … “What is the size of the EXISTING market”. Also important is how they ask it, the implication is that there better be a billion dollar market so they can do their financial calculations and plan that they can make 3% of the EXISTING market year 1 and do their normal left-brain quantitative forecasting yada yada yada.

    And so it stands that, because no big player with ability to execute has tried a trapeze-like solution and strategy, then none of them will. All of the buffoons are complete followers and it looks like now they will wait around for Cisco to take the lead in campus blowing a massive opportunity that would have set them up for a slam dunk as the technology matures enough for the data center.

Leave a Reply

You must be logged in to post a comment.

Login


MODAL