Times are tough all over. But, in an effort to see where “bright spots” still exist for service providers, xchange recently did a series of interviews with various luminaries.
Below is a Q&A based on an interview xchange Editor in Chief Paula Bernier did with Luc Ceuppens, senior director, head of product marketing for the High-End Systems Business unit at Juniper Networks Inc. (JNPR), and a standing xchange blogger. In it, Ceuppens discusses the opportunity for telcos to wholesale their IP networks to companies like Google.
xchange: In an effort to spread holiday cheer and help service providers identify opportunities for the year ahead I’m looking for guidance on what opportunities remain for service providers despite the grim economy. Your thoughts?
Ceuppens: I think there’s an opportunity for people that have large networks to sell or lease these networks to other providers as their infrastructure. This is something we started pushing now that we’re getting ready to launch our multichassis platform for the T1600, in combination with the control plane JCS 1200. You can actually consider that system as a platform that allows service providers to create networks for others and make these physically secure and physically separated, so they’re almost like hardware VPNs.
xchange: So it’s a wholesaling opportunity. But wholesaling what exactly?
Ceuppens: It’s like wholesaling an IP network. I’m sure you’re familiar with our concept of business model evolution. The challenge service providers have is how do I make money with all these content providers that use my backbone to sell their services, and I don’t really know it.
There’s an opportunity there to say ‘Hey Mr. Content Provider, it’s not always easy to guarantee that your services are receiving the best treatment through the backbone. I can actually give you your own network inside my network.’
xchange: So the telco or other network operator is offering QoS on the backbone? Or is it selling CDN services? Or is that all the same thing?
Ceuppens: I think ultimately they could be the same thing. QoS, guaranteeing QoS through the network, is very difficult. You need an enormous amount of signaling to actually guarantee a pipe of a certain size and a certain priority.
xchange: Right, but CIMI Corp. analyst Tom Nolle told me last month that revenue per bit has dropped so low that the difference between best-effort and QoS-based traffic for telcos wouldn’t amount to much in new revenue.
Ceuppens: I think the core of the problem is that even if a service provider talks about a best-effort network, which is what the Internet was always supposed to be, they don’t really let it be a best-effort network. They’re looking at ‘Oh, I’ve got 40 percent and 50 percent utilization on part of my network, I need to upgrade it.’ So they’re basically giving everybody the same treatment. That makes it very difficult to sell higher tier service because people say ‘I’m actually pretty happy with the best-effort stuff.’ So the first step would have to be that they let their best-effort network really be best effort and drop a packet every now and then.
