In the three years since the term NFV came into common parlance there has been a great deal of debate about the technology systems that can enable communications service providers to become more agile and efficient, and many discussions about the role of open source in network operator technology road maps.
Now we have reached crunch time. After many trials, proofs of concept (PoCs) and exploratory service launches, multiple major telcos are about to deploy New IP technologies -- SDN controllers, orchestration tools, virtual network functions -- as part of their commercial, production networks, as we saw this week. (See DT, Vodafone to Launch SDN-Based VPNs, SoftBank Turns to SDN for VPN Upgrade and DoCoMo on Track for Live NFV Deployment in Early 2016.)
But there's a problem.
Despite years of R&D by vendors, industry groups, open source projects and, it seems, anyone with a "Dummies Guide to Coding" in their back pocket, the operators can't get hold of all the technology building blocks they need to fulfill their visions. Sure, they can get hold of some of it, but there are important pieces of the New IP puzzle that are either unavailable (and so have to be developed from scratch) or they are not "fit for purpose" or "carrier grade," to use common terminology.
A very high profile case in point is OpenStack , which last week was publicly criticized by BT's chief researcher for data networks, Peter Willis. The BT Group plc (NYSE: BT; London: BTA) man gave a presentation at the SDN & OpenFlow World Congress that identified six key challenges that need to be addressed by the OpenStack community. (See BT Threatens to Ditch OpenStack.)
I think there's no doubt that the develop and vendor community will do its utmost to make OpenStack "carrier grade," but in the meantime, what are operators such as BT to do?
The British incumbent would much rather the OpenStack community fixed the issues it has identified, so that it doesn't need to develop its own cloud resource management system. The big debate in the coming months is going to revolve around how the OpenStack community can respond and whether any other non-proprietary alternatives are available and suitable for communications service providers.
If not, then it's also possible that proprietary solutions could become more attractive. But that's not what the New IP world is about, is it.
Or is it?
— Ray Le Maistre, , Editor-in-Chief, Light Reading