Back in my days as a magazine editor, I published a story about the supposed death of mainframes, wonderfully titled in a steal from Monty Python, “I’m Not Dead Yet.” And whadya know, mainframes still aren’t dead.
I was reminded of the whole idea of “not being dead yet” as I started looking at how carriers were making the move to SDN and NFV. I would have thought that the carriers would be wholeheartedly into the prospect of network virtualization. But while there’s a lot of announced commitment, there’s also some hesitation.
That’s not necessarily a bad thing. Carriers have invested a lot into creating highly reliable telecom networks, and no one would blame them if they didn’t want to declare those components dead yet.
Let’s start with the optimism, brought to us as always by the research firms. As Sean Michael Kerner reported in Enterprise Networking Planet earlier this month, network-focused Infonetics Research is forecasting big things for the combined SDN/NFV marketplace. What was a $500 million market in 2013 will blossom to $11 billion by 2018, in part because of telcos’ interest.
Not surprisingly, as most of the major telcos around the world have committed to SDN in one way or another. As we noted in our earlier post on standards, Telefonica, BT, and Verizon are all early supporters of NFV.
Down under, the major telcos are also deeply focused on NFV. As Stuart Corner noted in IT News Australia last week, both of the country’s major telcos, Telstra and Vodafone, are moving far beyond hype. “The goal of having all the dedicated hardware and software needed to operate a telecommunications network replaced by software running on standard servers - or 'network functions virtualisation' (NFV) - is proving to be the nirvana of the telecoms world.” Corner quoted Vodafone Australia’s CTO as saying that the telco will virtualize “most of its core network within five years.”
And yet, there are whispers … and sometimes even yells.
Dan Joe Barry wrote in Light Reading last week about the big advantage of SDN and NFV to carriers: “The flexibility and agility that SDN and NFV are intended to provide are expected to make it quicker and easier to configure services.”
No argument there, but Barry believes that the world has changed, with these capabilities “based on the assumption that carriers are in control of the services that are offered and consumed on their networks. It has a strong historical basis, as carriers have traditionally had this kind of control.” Now, argues Barry, the over-the-top (OTT) content providers own the services.
Frankly, I’m not sure why that’s a problem. If carriers can offer more services, no matter who they belong to, and consumers buy more them, everybody makes money.
But there’s more. Last June, a senior vice-president of Verizon said at a vendor conference that while his company is strongly committed to SDN and NFV, he’s concerned not only about hype, but also the “operational impacts of SDN and NFV.” At another conference last month, executives from both CenturyLink and Sprint also expressed doubts about capital expense reductions from moving to SDN. The CenturyLink executive was quoted as saying, “Telco central offices weren't built to house datacenter equipment and will need to be outfitted to do that, making the total cost of ownership (TCO) of a virtualized network about the same as today's capex budgets.”
And BT recently published a white paper (search on “BT 2014 Two TLAs” to download) in which it pointedly said, “We do not expect NFV to replace the core and access components of our global network. These rely on high performance custom hardware to deliver the performance and scale services our customers need of our services. The core connectivity between hundreds of nodes in over 170 countries serving hundreds of thousands of customer sites requires significant investment in fibre and physical infrastructure. This engineering masterpiece can never be virtualised.”
It’s clear that telcos are being cautious. Both NTT and AT&T are primarily looking at SDN in the context of providing cloud-based services to customers – not to necessarily use it within their own networks. Let’s be realistic here: Caution isn’t necessary a bad thing, especially among the folks who keep communications pinging around the world. Deep down, when it comes to reliability, the industry has always preferred to make computer networks more like traditional phone networks, not the other way around. Start in the shallow end, not on the high dive.
The real question is, what will make carriers move toward an unqualified commitment toward SDN and NFV? It’s going to take a concerted effort on the part of vendors to show them where such services fit into their offerings. For instance, consider this scenario put forth by Mitch Wagner in his Light Reading article about Nuage’s announcement last week of its Virtualized Network Services (VNS) targeting carriers: “Instead of sending out a technician to install hardware equipment, load software and configure it, carriers can just drop-ship a box to a customer site and activate service remotely.”
Wagner also noted, “VNS is the very essence of what carriers are looking for in SDN and NFV -- the ability to get past expensive proprietary hardware requiring manual installation and configuration to instead deliver services over the network, automatically, without manual labor.”
There’s a strategy. When targeting carriers, vendors can’t advocate a rip-and-replace; no one in their right mind would do that. There needs to be a middle ground to show carriers where the advantages of SDN and NFV lie, and Nuage’s strategy may just show the way.