The Big Telecom Switch
I was expecting to be crazily busy in the post-purchase phase (expectation met!), but what I wasn't prepared for was just how much our industry had changed while I was away. (For those unfamiliar with my "backstory," I took six years off from telecom from 2007 to 2013, either running an integrated marketing agency or serving hard time for criminal negligence, depending on your choice of grapevine.)
Coming back was a bit like being cryogenically frozen -- Han Solo-style -- and then being thawed out and finding that just about everything you thought you knew about telecom had changed.
To help me get an up-to-date handle on what's happening today, and to avoid making a giant ass of myself in front of my old/new Light Reading employees, I undertook what all new CEOs are supposed to do -- a series of 100 customer meetings in my first 100 days back "in office."
These meetings were mostly with C-level and Senior VP level execs, at both service providers and equipment manufacturers, and under NDA, a condition that probably provided me with a more accurate indication of what the industry really thinks than the paid marketing messages now competing for your attention in the html pages of industry websites.
So what did I learn on my journey through the telecom industry?
By far the biggest change evident since I left is the shift in priorities amongst Light Reading's service provider audience.
In 2007, network operators were still entrenched in the business of building high-capacity networks and making sure they were reliable. And that work continues, of course, every day. But today's service providers now face a new challenge -- monetizing all that capacity, and doing so in a way that both maintains acceptable levels of profitability while differentiating them from a slew of competitors, old and new.
In fact, it seems to me that it might be time to revisit and update our old friend the OSI model by adding an extra layer -- Layer 8, or the Money Layer, as shown below in a slide from my keynote presentation at the recent Big Telecom Event. Because that's really where the focus of the industry is right now: this segue from building the network to monetizing the network.
(Note: It's apparent that as the industry's interest moves up into the higher layers of the model and beyond, so the level of interest in lower OSI Layers such as 1 (optical), 2 (switching), and 3 (routing) is decreasing. If not actually commoditized, there is certainly far less to talk about or debate in areas such as carrier class Ethernet or IP routing than there was 10 years ago, something which raises all sorts of questions about the future role and relevance of organizations such as the Metro Ethernet Forum (MEF) .
The most obvious effect of the monetization trend is the way it is now driving the world's telecom providers to undertake a dramatic business transformation process that is manifesting itself in a number of different ways: mobile operators looking to compete with terrestrial operators; cable operators looking to move into the mobile market; OTT operators buying fiber assets; and Tier 1 operators trying to beat the newcomers at their own game.
Factor in new players such as municipal and county governments, broadcasters, and utilities, and things get really interesting. (Obviously, these are all sectors that both Light Reading and Heavy Reading will be covering intensively during the next year, so stay tuned for lots more on all of these markets).
So that's the big picture. But within that superset there were a number of noteworthy technology trends that kept recurring in my meetings earlier this year.
The most notable, by far, relates to SDN and NFV. This is obviously the story of the year -- maybe the decade. And it's clear that of the two, SDN has so far swallowed the majority of marketing budgets and publicity. (Why, some people are even naming their industry publications after SDN!)
But behind the scenes it's NFV, not SDN, that dominates the conversation and the excitement.
Why? Because NFV is what service providers have decided they want, and need, right now. And that is a unique and unprecedented occurrence in the history of the telecom industry. For the first time, we are seeing the top technologists in the service provider community get out in front of the vendors and set a new direction (and, it must be said, catching some of the incumbent equipment manufacturers on the hop).
So what on earth is going on with SDN? Is it possible that our entire industry has just gotten it wrong on this tech?
It's happened before, after all, as those of you (like me) that are old enough to remember the battle between ATM and IP in the late 1980s will recall. Back then pretty much the entire trade press backed asynchronous transfer mode to wipe IP and Fiber Channel off the network. Thirty years later and, well, you know how that worked out.
But that kind of catastrophic error of analysis isn't what is happening here, I think. There's no question that 21st century telecom networks must be automated and virtualized, but the reality is that defining SDN and implementing it in wide area commercial communications networks is non-trivial.
Regardless of what industry pundits think, the process will take years -- probably a decade, in fact.
What type of SDN will emerge at the end of that time? Forget about any SDN solution that requires service providers to rip and replace boxes to obtain benefits. That's not going to happen.
Overlay solutions offer advantages, but isn't SDN supposed to be about simplification and consolidation?
As has happened many times before, it's likely that a hybrid will emerge, with networks that are more programmable than they are today but still recognizable as wide area communications systems.
What's intriguing -- and where companies stand to win or lose an awful lot of money -- is working out how our industry will get from where we are now to that new architecture. And in the meantime, NFV offers immediate benefits.
So those were the biggest findings of my tour of telecom during the first quarter of this year. Do you agree with them? Please let me know on the message board below.
— Stephen Saunders, Founder and CEO, Light Reading