Our SDN Call to Action

March 15, 2016
By John Donovan

Network trends are staggering and not slowing down. As Andre Fuetsch noted recently, we’ve seen more than 150,000% growth in mobile data traffic since 2007 and that’s expected to grow ten times by 2020. 60% of our total network traffic is video. We see more than 114 petabytes on our network daily. That’s equivalent to streaming almost 130 million hours of HD video in one day.

These facts and predictions have driven us to re-architect our network based on a new paradigm, one where our ability to scale is based more in software than in the hardware. I’m at the Open Networking Summit in Santa Clara today to talk about the progress we’ve made and where we’re going. Last year, I described how we’re using CORD to accelerate our roll out of GigaPower. As promised, our trials will start in the next few weeks. We’re also all in on virtualizing our network.  Our goal is to get to 75% by 2020.  Last year, we committed to 5% in 2015, laying the foundation.  We beat that at 5.7%.  We plan to accelerate that to 30% in 2016.  As of today, 14 million wireless customers are on our fully virtualized mobile packet core, with millions more being migrated in 2016. These are examples of how virtualization touches every part of our network transformation.

One of our biggest challenges was that there was no playbook here for us to follow when it comes to virtualizing and software-controlling a Wide Area Network. But the technical and economic principles are sound. So, we decided to take the lead in writing the playbook for NFV/SDN deployment in the WAN. And we’re excited to start opening that playbook to developers as well as connectivity and cloud providers.

For almost two years now, we have been architecting and coding a large software project called ECOMP. That stands for Enhanced Control, Orchestration, Management and Policy. It’s a mouthful. But it’s also important. ECOMP is an infrastructure delivery platform and a scalable, comprehensive network cloud service.  It provides automation of many service delivery, service assurance, performance management, fault management, and SDN tasks.  It is designed to work with OpenStack but is extensible to other cloud and compute environments. ECOMP is the engine that powers our software-centric network.

Now, we’re opening the hood of our network and showing you the engine. ECOMP automates the network services and infrastructure that will run in the cloud. A system like ECOMP is very powerful as it allows us to build our next generation cloud-based network in a vendor agnostic way, giving us great flexibility for deploying NFV / SDN in our network. As a model-driven platform, this framework costs less than maintaining existing network systems. And it allows us to accelerate the implementation of new services quicker than ever before.  ECOMP is one of the most challenging, complex and sophisticated software projects in AT&T’s history.

So, what’s next? We have written a whitepaper on ECOMP that we’re making publicly available starting today.  We did this to give the industry an idea of our thinking and direction. On a global scale, we know the needs we have are similar to the rest of the industry and other cloud services providers. You’ll hear more about this from Chris Rice, VP of AT&T Labs in the next few days.

As a start, we would like to hear from others in the cloud community about their thoughts on our vision. We are so committed to this vision that we are amenable to releasing ECOMP into open source.  Where we go from here, however, depends on the feedback we get from the cloud and developer communities. We believe that one of the most important tenets of the open source community is that you don’t just take code. You contribute it, as well.  This collaborative spirit is a driving force in our transformation.    

Download the paper. Tell us your thoughts. We hope you’ll want to be part of how this all unfolds. 

Filter by:

  • Most Popular
  • Author
  • Tags

Archives

Link to archives