So, we wrapped up our day with the Networking Field Day crew last week with a free form discussion on where we go next with SDN. To be honest, the session did not go quite as I envisioned, but, in retrospect, I would not changed anything. As Ethan Banks (of PacketPushers fame) noted in Twitter, this session was more about shooting the unicorns than letting them run free. It seems that if we are going to convert our SDN unicorns into SDN plough horses, we are going to shed a little blood. At the end of the day, the market will be served by frank conversations—we need to move beyond painting SDN acolytes as starry-eyed and SDN detractors as being heretical and reactionary.
In the interest of keeping the conversation going, here are some of the things I walked away with after the conversation on Wednesday (in no particular order):
Is Hardware Innovation is Over?
This industry has always been one big pendulum and, currently, the pendulum is firmly in the software camp. Today, many of the truly interesting things in networking are going on with software. While most would agree we are at an inflection point with programmability, there are no clear directions for the evolution of SDN. Certainly there are pieces in place like OpenFlow and OpenStack, but OF 1.3 in unlikely to be the zenith of OF evolution let alone SDN evolution—current technologies will continue to mature and new ones will inevitably emerge. More importantly, the “how we do things” and “what do want to accomplish” of SDN will most certainly continue to evolve and as long as that is the case, software will rule because it’s simply easier and faster to experiment with software. But, once some clear directions begin to emerge, I guarantee you the action will swing back towards the hardware because doing things in hardware tends to be faster and more efficient. I could point to Cisco examples of this, but instead look at what Intel, the poster child for general purpose processors, has done with VT extensions to support virtualization or QuickSync for video transcoding.
Is OpenFlow Ready for PrimeTime?
One of the more contentious points yesterday is if OpenFlow is production ready. I think it’s a flawed “do these jeans make me e look fat” kind of question. There are certainly folks out there using OF to handle production traffic—for example, some of the cool things Brent Salisbury is doing. So, it’s not a binary question, but more a matter of assessing scope and scale. The better question to ask is what is the operational and performance envelope of OpenFlow and how does that match my needs, priorities, and capabilities. The risk with any emerging technology is that, often, the only way you find the edge of the envelope is once you’re on the other side, usually with colorful and memorable results. Regardless, I don’t see this question existing in another year or so.
What better way to spend Valentine’s day than to watch a webcast on OpenFlow and SDN, perhaps with your significant other? The last couple of years have seen considerable buzz around aspects of software-defined networking. A significant portion of the early seed discussion was around OpenFlow. As part of the Cisco Open Network Environment webcast series, this time on February 14th, 2013 at 9 AM PST, we take look at an :Introduction to OpenFlow”: What is it? How does it work? What are some of the potential use-cases?
Joining me in this discussion with be David Ward, Cisco CTO of Engineering and Chief Architect. At the time of recording David also wears the hat of the being the Chair of the Technical Advisory Group at Open Network Foundation (ONF). So he brings perspectives both as someone who’s driving the evolution of the protocol, as well as somebody guiding its implementation across several products within the Cisco portfolio.
Also joining the webcast to lend end-user perspectives will be Matt Davy, who is formerly of Indiana University, having been the executive director of the INCenter facility there. Matt’s recently moved onto a new role, but he built a lighthouse test bed around OpenFlow and SDN the last few years during this employment at the university. Matt will talk about campus slicing and his experiences around OpenFlow. Providing service provider perspectives from NTT communications will be Yuichi Ikejiri, Director of the Network Technology Services division.
As mentioned before, this is part of an educational series. If you’ve not watched the first in the series, entitled “An Introduction to OpenStack” – please feel free to register and watch it here. The panel of Lew Tucker and Raj Patel below provide interesting perspectives on OpenStack.
Topics such as Software Defined Networking and programmable networks are of great interest to many network operators these days. With that in mind we’re pleased to kick off a new series of Webcasts to help our customer learn more about Cisco’s Open Network Environment strategy.
Please join us on February 14, 2013 (9:00 am PST, 5:00 pm GMT) for an educational webcast featuring David Ward, Cisco CTO, Engineering and SP Chief Architect, and Chair of the Technical Advisory Group at the Open Network Foundation (ONF). It’s a chance to learn more about OpenFlow and its components, the protocol evolution and use cases, as well as its integration into existing networks. Read More »
Consider these impressive stats shared in a keynote from Cisco’s CTO and CSO Padmasree Warrior last week at Cisco Live, London:
50 Billion “things” including trees, vehicles, traffic signals, devices and what not will be connected together by 2020 (vs. 1000 devices connected in 1984)
2012 created more information than the past 5000 years combined!
2/3rd of the world’s mobile data will be video by 2015.
These statistics may seem a bit surprising, but the fact is, they cannot be ignored by CIOs and others chartered with the responsibility of managing IT infrastructure.
Impact on Enterprise and SP Infrastructure strategies
Further, these trends are not silo’d and are certainly not happening in a vacuum. For example, Bring-your-Own Device (BYOD) and the exponential growth of video endpoints, may be happening in the “access”, but they are causing a ripple effect upstream in the data center and cloud environments, and coupled with new application requirements, are triggering CIOs across larger Enterprise and Service Providers to rapidly evolve their IT infrastructure strategies.
It is much the same with cloud infrastructure strategies. Even as Enterprises have aggressively adopted the journey to Private Cloud, their preference for hybrid clouds, where they can enjoy the “best of both worlds” – public and private have grown as well. However, the move to hybrid clouds has been somewhat hampered by challenges as outlined in my previous blog: Lowering barriers to hybrid cloud adoption – challenges and opportunities.
The Fabric approach
To address many of these issues, Cisco has long advocated the concept of a holistic data center fabric, heart of its Unified Data Center philosophy. The fundamental premise of breaking silos, and bringing together disparate technology silos across network, compute and storage is what makes this so compelling. At the heart of it, is the Cisco Unified Fabric, serving as the glue.
As we continue to evolve this fabric, we’re making three industry-leading announcements today that help make the fabric more scalable, extensible and open.
Let’s talk about SCALING the fabric first:
Industry’s highest density L2/L3 10G/40G switch: Building upon our previous announcement of redefining fabric scale, this time we introduces a New Nexus 6000 family with two form factors – 6004 and 6001. We expect these switches to be positioned to meet increasing bandwidth demands, for spine/leaf architectures, and for 40G aggregation in fixed switching deployments. We expect the Nexus 6000 to be complementary to the Nexus 5500 and Nexus 7000 series deployments, and is not to be confused with the Catalyst 6500 or Nexus fabric interconnects.
The Nexus 6000 is built with Cisco’s custom silicon, and 1 micro-second port to port latency. It has forward propagated some of the architectural successes of the Nexus 3548, the industry’s lowest latency switch that we introduced last year. Clearly, as in the past, Cisco’s ASICs have differentiated themselves against the lowest common denominator approach of the merchant silicon, by delivering both better performance as well as greater value due to the tight integration with the software stack.
The Nexus 5500 incidentally gets 40G expansion modules, and is accompanied by a brand new Fabric Extender – the 2248PQ, which comes with 40G uplinks as well. All of these, along with the 10G server interfaces, help pair the 10G server access with 40G server aggregation.
Also as part of the first step in making the physical Nexus switches services ready in the data center, a new Network Analysis Module (NAM) on the Nexus 7000 also brings in performance analytics, application visibility and network intelligence. This is the first services module with others to follow, and brings in parity with the new vNAM functionality as well.
Industry’s simplest hybrid cloud solution: Over the last few years, we have introduced several technologies that help build fabric extensibility -- the Fabric Extender or FEX solution is very popular extending the fabric to the server/VM, as are some of the Data Center Interconnect technologies like Overlay Transport Virtualization (OTV) or Location ID Separation Protocol (LISP), among others. Obviously each have their benefits.
The Nexus 1000V Intercloud takes these to the next level by allowing the data center fabric to be extended to provider cloud environments in a secure, transparent manner, while preserving L4-7 services and policies. This is meant to help lower the barriers for hybrid cloud deployments and is designed to be a multi-hypervisor, multi-cloud solution. It is expected to ship in the summer timeframe, by 1H CY13.
This video does a good job of explaining the concepts of the Intercloud solution: