With two bold moves in as many weeks, AT&T is making its commitment to open source very clear. And it is also making clear the challenges it sees to achieving the ambitious goals set to virtualize 75% of its network functions by the year 2020.
First, AT&T Inc. (NYSE: T) shared its vision for automating orchestration and network management, known as Enhanced Control, Orchestration, Management and Policy or ECOMP, through a whitepaper, and indicated a willingness to share the 8 million lines of code it has developed as an open source project. (See AT&T Shares ECOMP Vision, Might Share Software.)
That announcement came at the Open Networking Summit. A week later at the Optical Fiber Conference, AT&T again launched a new initiative, this one aimed at bringing software-defined networking (SDN) down to the optical layer. The Open ROADM project thus far includes three vendors but no other service providers. (See AT&T Targets Virtualized ROADMs.)
In each case, the purpose of sharing information and even publishing specifications for Open ROADM was to engage the broader telecom community in an effort that one premiere telecom operator thinks is vital. AT&T is not alone in doing this -- Telefonica, BT, China Mobile and China Telecom have all been engaged to some degree in efforts to rally the industry around what they see as critical orchestration efforts. And operators have come together multiple times, once to launch the NFV process through ETSI, and later to create a group that would address NFV infrastructure through the Linux Foundation.
What makes AT&T's efforts a bit different is that the company is willing to stand alone, at least initially, and point firmly in a direction it believes the telecom industry needs to turn. What gets interesting, from this point forward, is the industry's reaction to what AT&T is offering, and the direction in which it has chosen to go. (See AT&T Stresses Its Broader NFV Vision.)
At least one optical industry veteran, who was at OFC last week, admitted to me that there was some question by others at the show about whether there's value in driving virtualization down to the physical layer of the optical network. Andre Fuetsch carefully explained to me where AT&T sees that value, and it's in making the physical network as flexible and software-controlled as the layers above it. (See AT&T Embracing SDN at the Optical Layer .)
Will others be persuaded? We'll have to wait and see. Clearly, AT&T has its own reasons for wanting to line up industry support behind initiatives it considers critical, the number one being additional momentum to move things forward faster. I don't think you can fault their approach. The question is, will it succeed?
— Carol Wilson, Editor-at-Large, Light Reading