Network functions virtualization (NFV) is now such a crucial development to the future of communications network operator strategies that it's somewhat surprising to recall that, in effect, it's less than two years old.
And look where we are today: NFV permeates nearly every corner of our industry, from the chip level up. This is just one of a number of examples that tells us something about the way the telecom sector is changing. The legacy technology calendar, whereby new concepts took many years (up to a decade in some cases, maybe longer) from an initial concept to commercially available product/application/service, is no longer relevant. Those days are over for anyone that wants to be a survivor and innovator in today's networking sector.
Baby, you've grown up fast!
I was lucky enough to be around when a group of senior operator executives gave birth to NFV in Germany in October 2012, and was lucky enough to be in the delivery room – that is, when the ETSI NFV ISG (Industry Specifications Group) was announced and the first co-authored, operator-led NFV white paper was waved in public.
Obviously the group had been engaging for some time and the white paper had its gestation period, but it was on October 23, 2012, in Darmstadt, Germany, when Deutsche Telekom network architect Uwe Michel stood in front of the attendees at the SDN & OpenFlow World Congress to announce the formation of the ISG and to unveil the influential white paper. (See Carriers Collaborate on Network of the Future.)
At that point the group comprised 13 operators, and was at the initial stage of identifying the opportunities and challenges/pitfalls that NFV would deliver. And at that point, the idea was to formally initiate the ETSI group and give it a two-year life to determine whether NFV was a useful concept or just a good idea with too many challenges.
Now, less than two years later, the NFV ISG has more than 90 members (operators and vendors) and about 200 additional "participants" (mostly vendors).
And its two-year shelf-life has been extended – the response to the concept and the support it received from all corners of the telecom network infrastructure ecosystem meant it had to continue the work it had started.
The group has done more than just sign up a lot of new members and given itself an extended term. Following a somewhat turbulent start as the various members "debated" the best way to achieve tangible results, the group has spawned 23 proofs-of-concept (PoCs), based on the NFV ISG's Proof of Concept Framework, all of which feed their results and findings back into the broader group to help determine the best way to develop technologies, deployment plans and business cases.
Operators and vendors are working together on these PoCs, often with open standards tools at the heart of their developments, and there are plans for NFV test beds being run around the world by various operators to be federated. Virtualized functions are being introduced into production telecom networks as we speak (read?), with 2015 set to be a breakout year for the deployment of all manner of tools that will support large scale NFV deployments.
Let's step back for a moment and consider how long all this would have taken if a new concept had been introduced in, for example, 2001. My guess, based on a mixture of observational experience along with a dash of cynicism, would be about 6 or 7 years – that's the kind of timescale that comes with traditional standards bodies processes, test processes, planning, marketing and corporate strategic decision-making. And procrastinating. And then deciding that "Hey, that doesn't really need to happen just yet, so why bother?"
Actually, now that I've written that down, let's revise that to 8 or 9 years.
Torn up and the pieces scattered
The point is, the old rule books have been torn up. The traditional standards bodies and industry forums that for years have determined the pace of change in wide area networking are no longer the overlords. Traditional telecom vendors and operators are no longer the companies that determine R&D, product development and market introduction timetables -- face it folks, the ecosystem is now being forced to work at a pace (and in a way, operationally) by the drum beat of the Web services giants such as Google and Amazon: They won't wait for a three-year ratification process. Those days are over.
For the telecom community, the pace at which NFV has developed from a multi-page white paper to a hotbed of collaborative, real-world development is an indication of how the sector is adjusting and growing up, and an indication that everything has changed.
And most of the telecom ecosystem's participants have realized this. In a way, the industry is currently in a state of what might be called controlled chaos, albeit rather muted as chaotic situations go. But while you may not be able to hear strategists, marketing managers, and every other person involved in deciding "what comes next" scream out loud -- though maybe you can, depending on where you're sitting -- there's a lot of soul-searching and upheaval underway, and it's helping to change the shape of the industry.
The question is -- who can keep pace and embrace the new world?
In many ways, it's just a matter of time.