Scott Puopolo, VP and Global Head of Cisco’s Internet Business Solutions Group (IBSG) Global Service Provider Practice, discusses the progress being made within the wide-ranging CDN federation trials announced last year.
Can you tell us more about Cisco’s CDN Federation Pilot for service providers?
Sure. This actually all kicked off at last year’s CDN World Summit (although we’d been planning it for months before that), when we announced the three-phase pilot, which was a first-ever experiment involving CDN linkages between British Telecom, KDDI, Orange, SFR and Telecom Italia.
We did it because federating CDNs and optimising them for video is critical to keeping up with worldwide device proliferation. All these new screens want video! Lots of video.
Yes, service provider CDNs play a unique role in enabling a high quality of experience (QoE) for rich media services, like video – but if they’re not interconnected, they won’t go as far as they could.
By federating CDNs, service providers lower costs, by pooling resources, and content producers reduce the business and technical complexities, simply by dealing with fewer companies. Ultimately, when we are successful, consumers benefit by getting better QoE/QoS on all of their video screens.
Second, a technical assessment, to identify CDN architectures, capabilities, approaches, and roadmaps for activities like request routing, content distribution and accounting/reporting.
Third, a lab trial, to validate and test the notion of an open CDN federation.
And the effort isn’t just contained to the pilot, by the way. My colleague Francois Le Faucheur is one of the life-forces behind the CDNi initiative (where the “i” stands for “interconnection”), an active IETF effort. You’ll be hearing from him as well, during the run up to this year’s CDN World Summit.
What progress has been made so far in this initiative?
Oh, lots! But if I tell you all of that here, what will I tell you during my keynote this October? I’m kidding. Mostly. Here’s what I can say at this point: This is a 3-phase effort.
Phase 1 you know, or you do if you were at last year’s CDN World Summit. The history of it is, we at Cisco kept hearing from our content and service provider customers about a problem set that was so similar, one to the next, that we opted to act as a catalyst – by introducing the notion of the open and federated CDNs.
Soon we’ll be discussing specific results from phase 2, and who is new to the party since last year. And maybe a bit about what’s on tap for phase 3.
What have been the program’s major sticking points so far?
Getting multiple service providers and content owners to agree on business and technology techniques – what could possibly go wrong?! Any effort involving multiple parties, some of them competitors, encounter challenges along the way. Ours is no different.
I’m happy to report, however, that the majority of them fall into the category of “good problems to have.” One that comes to mind is the many different “flavours” of federations that could happen – local vs. intra-carrier, as one of many examples.
A lot of people tend to think of a federated CDN as this big, monolithic, global thing. And that’s surely one potential instantiation, but we’re also seeing a lot of interest from local and regional players in federated CDNs.
Each one is a different flavour – and we think those different types of instantiations will proliferate. Different flavours means different business models, and potentially different technological approaches. So that’s been a learning process.