What’s the foundation of the next generation data center network, i.e. this thing some call the software defined virtual data center network?  Many companies have recently re-branded their products and jumped on the Software Defined Networking (SDN) bandwagon in some way, shape, or form, and for good reason.  It has the potential to truly change networking as we know it today.  IDC has even stated SDN could be a $2B market by 2016. 

Let’s forget about all of this recent SDN washing and go back to virtual networking basics.  Most of us by now know what a software switch is.  It is also known as a vswitch or virtual switch.  This is arguably the most critical piece of real estate in the next generation data center network.  So, who owns this property?

For those who visit here often know I'm a bit of an SDN purist , maybe even a Revolutionary.  But this article, it stemmed from several recent conversations on how the network will EVOLVE over time.  This pertains to data center only, so it's more relevant to the topic of *Empowering* the Software Defined Data Center (SDDC).  It was also meant to be a high level quick start guide regarding these trends.  For more details, I've written other posts and there are definitely a lot of others out there too. 

Here it goes.

Network Virtualization & Software Defined Networks

As an infrastructure and IT professional, you are probably overwhelmed with the amount of information and hype right now regarding topics such as OpenFlow and Software Defined Networking (SDN).  Rest assured you are not alone.  My goal is to simplify this messaging, review what can be done in the world of software and virtual networking today to meet the demands of Cloud, but will also lay the foundation for the next generation network.

There is no better time than now to be in the world of networking.  While it is changing significantly without many taking notice, we have exciting times ahead of us.  Many of us, including me, may even be out of a job in a few years once networking becomes truly automated, but for now, let’s embrace the change and see what happens!  Just in the past few months we’ve seen Nicira, vCider, and now Vyatta get acquired, not to mention the other SDN startups getting more VC funding, the most recent of this bunch, Big Switch Networks.  But, today’s announcement is clearly about Vyatta getting acquired by Brocade.

It was just announced Riverbed will be acquiring OPNET.  With the growth of BYOD, Cloud, SDN, and Collaboration just to name a few of today’s hottest trends, it is now more important than ever before to have deeper visibility into both the network and the applications riding over the network.  For the mid-size Enterprise in my experience, they usually rely just on SNMP, WMI, and sometimes NetFlow to gain visibility to the network.  However, this data on its own is not enough to really know what’s going on throughout the network.  From my perspective, network and application performance management (APM) solutions are those that the incumbent network vendors should have been selling for the past decade.  These are what’s really needed.  How can you make a better network or make applications run smoother if there isn’t direct integration between the network and the applications (via an APM tool)?

If we use last year’s Interop as the OpenFlow/SDN coming out party, it took just over a year for Cisco to fully develop and announce a comprehensive multi-segment strategy.  Their SDN encompassing strategy is called Cisco Open Network Environment (ONE).  Congratulations, Cisco!  If they got David Ward back from Juniper sooner, maybe the strategy would have already been announced.  Joke…I really don’t have any insight as to who was or is responsible for the strategy, but would imagine it to be a fairly extensive team.

I think it was a good move to announce during Cisco LIVE.  Customers worship Cisco, not just for the products, solutions, architectures they develop, but also for this week long party where they receive gifts and gadgets, and soak up some of the most technical content in the industry, but most importantly can be around like-minded individuals.  That is the most important thing for those that are technically inclined and is often not understood by those who aren’t “down in the weeds.”

Greg Ferro does a nice job here directly stating the networking incumbents should step up with an SDN strategy.  I agree 100%.  Brad Casemore also chimes in with his thoughts.  If you aren’t already reading their blogs, I encourage you to do so because you’re missing out.

Several companies have announced they have OpenFlow-enabled switches, but for these companies, there is still no strategy and no reasoning as to why their switch should be used when deploying an OpenFlow based SDN.  Furthermore, they lack a strategy overall looking at the various components of a Software Defined Network.  From a hardware standpoint, some of the same features and characteristics  (buffers, table sizes, etc.) will still need to be compared as we already do today in traditional networks, but even that, isn’t documented in these announcements.  A lot of these vendors think they are on the offensive [vs. Cisco] announcing OpenFlow enabled switches (without a controller), but they really aren’t, in my opinion.

[Before you start reading, I need to give a big thanks to  Christian Esteve Rothenberg, Research Scientist at CPqD.  I asked if he wouldn't mind reading through this post prior to posting to ensure I didn't botch up anything on RouteFlow, and sure enough he immediately helped out and provided great feedback.  Christian also provided us with the picture you'll eventually see below and many of the RouteFlow links as well.  There was much more information he provided that I'll hope to get out soon too.  Thanks again, Christian.]

So...let's get to it.

Understanding “flow based protocols” and RouteFlow can change the way you think about networking and the protocols we use on daily basis.  I’m referring to control plane protocols such as Spanning Tree, OSPF, EIGRP, and BGP.  Based on the traction from this blog, I can see many people are searching for answers on what SDN means for the industry and what the future will be for a network engineer, etc.  If you are one of those people, first, ready my last post. It’s a quick synopsis of a presentation at ONS that covers some interesting automated tools already available for controller based networks.  And second, keep reading here.

I mentioned in my previous post Nick McKeown, uber smart Entrepreneur and Professor at Stanford, gave what I thought was the finest presentation of the week at the Open Networking Summit hosted in Santa Clara this week.  As I wait to board my plane back to the East coast, here is a more detailed recap of the presentation and what I took away from it…

Here is a quick summary on what I think worked well, what didn't work, and some thoughts on improving ONS next year.

What worked?

• Getting everyone together that has an interest in OpenFlow/SDN.  It was great to collaborate and talk with like minded individuals even though we, as an industry, are probably still on the incline of the hype cycle.
• Google, Yahoo, etc.  Anything these hyper scale companies talk about is damn interesting
• Exhibits – these are great for any conference and will only improve at ONS as more products are further developed and created.
• VC presentation by Lightspeed.  Great presentation here and I’m sure it got the wheels spinning for anyone with an entrepreneurial mind since they are still looking to invest and also believe it’s early in the life of SDN.
• Nick McKewon – see my next blog post.  This was the best presentation of the week.  Wish I could have seen Nick present all day.
• Pragmatic view by mostly all admitting there is still a way to go in this shift to SDN.
• Logistics, overall schedule, multiple projector screens, and the like. 
• There were many more presentation that were also intriguing, but those listed above were the ones top of mind right now.
• Overall, extremely glad I had the opportunity to attend ONS 2012.

With just a few minutes to spare until the 5:30 start of the evening event and exhibits, I thought I’d give a really quick summary of Day 1 at the Open Networking Summit 2012.  Note there were two tutorial sessions today and I attended the one for engineers.

The first thing you noticed by seeing everyone’s badges/name tags with associated company, and was confirmed by Brandon’s presentation in the first slide, was there was and is a truly broad audience here.  There are the obvious participants from the big name manufacturers, but also, there are between 1-3 people from at least 60 “other” companies, which is the category I fall into since I’m the only one representing BlueWater from NY/NJ.  I also had the pleasure of sitting next to the sole person from Aruba Networks as well.  Not sure what that tells you about their SDN strategy.