NEWS ANALYSIS: AT&T's Domain 2.0 initiative shows that the telecom company is serious about embracing software-defined networking.
Flexible, innovative and ready to lead the technology pack have never been terms that I have applied to AT&T—but I may change my mind.
I've been following the Open Networking Summit keynotes online—which are very well done; please, other conference organizers take note—and AT&T's embrace of software-defined networks (SDNs) is already one of the top networking stories of 2014.
The rise of software-defined networks has been championed by vendors and academics, but it really needed one big customer with deep pockets to come out and say, "This is the way we will build our network." At the ONS, the role of the deep-pocketed customer was ably undertaken by John Donovan, senior executive vice president of technology and network operations at AT&T.
"We're changing everything," Donovan said. That everything includes the networking architecture, the skills and the basic philosophy underlying the telecom giant's digital network.
Those changes were first outlined last September when the company announced its Domain 2.0 initiative aimed at rethinking the company's networking infrastructure and orientating toward software-defined networking and network function virtualization. Taken together, those two concepts seek to decouple the hardware network elements from software management.
The end result would be similar to cloud computing concepts where scale, flexibility and capacity are fluid services built around demand and customer applications. While that cloud concept has become something of the de facto expectation in computing services, the networking world is just starting to undergo a similar change. The current state of corporate and telecom networks grew up around tightly integrated hardware and software stacks provided by one vendor.
Donovan outlined four essential elements of the AT&T networking goal. Those four key principles included an open architecture, a simple model, scalability that can evolve with customer needs and security.
In early February, the company announced the first vendors selected to work on the Domain 2.0 project, which included Ericsson, Tail-F Systems and Metaswitch Networks. Missing from the current list, but not precluded in the future, includes the giants of telecom network vendors Cisco and Alcatel Lucent.
Remember, AT&T also recently said it
will boost capital spending to about $22 billion per year over the next three years as it upgrades its wireless and wireline capacity. Not all that money goes to the networking innards. But AT&T's stance will accelerate the move toward software-based networks and functional virtualization in a far faster way than what happened in the compute and storage segments of cloud computing.
Right now software-defined network is still in the definition phase, with standards organizations being formed, alignment with other aspects of the open computing stack under way and the traditional networking vendors realizing that if they don't get on the bus, they are going to be left behind.
Under Domain 2.0, the company intends to cut by two-thirds the time it takes to qualify new technologies for use in the network. This means the amount of time required to qualify new products would be reduced to a matter of months instead of years, meaning the qualifying process would require less time than the product development process for many traditional vendors.
AT&T (before it was broken up by a consent decree in 1982) was renowned for the technology innovation produced by Bell Laboratories, which is now part of Alcatel Lucent. The list of innovations from Bell Labs reads like a history of innovation in the technology industry and includes the development of the transistor, radio astronomy and the foundation of most modern programming languages.
Domain 2.0 represents a different kind of innovation that leverages the technical developments and new companies emerging from the software-defined networking model to inject innovation into the telecom networking infrastructure. This change is much needed and will be instrumental in setting the next era of networking not just in telecom but in corporate computing as well.
Eric Lundquist is a technology analyst at Ziff Brothers Investments, a private investment firm. Lundquist, who was editor-in-chief at eWEEK (previously PC WEEK) from 1996-2008, authored this article for eWEEK to share his thoughts on technology, products and services. No investment advice is offered in this article. All duties are disclaimed. Lundquist works separately for a private investment firm, which may at any time invest in companies whose products are discussed in this article and no disclosure of securities transactions will be made.