Carrier Ethernet: Beyond the Metro

Recently, I was involved in a panel presentation with experts from various segments of the industry to discuss those factors that make for successful adaption of Carrier Ethernet, which has been labeled the “natural choice” for deployment and how service providers are looking to increase Ethernet expansion within their networks as both a business value and technology migration decision. From that discussion and my numerous interviews with service providers, vendors, and equipment manufacturers, I have narrowed down the success requirements to two: scalability and addressing some of the service provider pain points.

Within the Metro network, CE is a value proposition. However, when going outside of the network, there seems to be a need to expand Ethernet with existing customers as opposed to maintaining legacy systems. Scalability and addressing some of the pain points the providers are facing are two main requirements of successful CE deployment. Just because it’s Ethernet doesn’t mean it is natural deployment; it has to be true carrier grade.

Ethernet evolution has reached the point where capability is now defined as being similar to layer 2 and layer 3 capabilities. For CE convergence to be effective it must have speed or service velocity within a network. Additionally, SPs need to be able to deploy the services across their entire network and offer multiple services. However, these are not the only factors that make convergence successful. Equally as important is enhancing the customer’s experience with these services. One of the most difficult pain points for SPs is the customer acquisition costs. Capturing and keeping customers is a top priority not only from a CapEx/OpEx perspective but from a growth perspective—being able to offer that customer more services and increase ARPU per user. But the bottom line is that none of these pain points matter unless the economics of the network or delivery within the network decreases.

The equipment being deployed outside of the Metro is over 17% CAGR—clearly there is growth in that segment. There are two different components associated with this growth. Carrier Ethernet acting as aggregation is the traditional layer 2 that has QinQ Ethernet over MPLS, pseudo wire, native IP, and MPLS VPN. But to deliver outside of the Metro you need to deliver new, enhanced services with the traditional layer 2 but with a QinQ that is able to be delivered into Ethernet over: MPLS, VPLS, layer 2 access into layer 3, native IP, MPLS VPN, encrypted VPN, IP security to deliver better application awareness, and the traditional VPLS, RFC2547 and Ethernet subscribers.

When you consider what this means to the service provider from an Ethernet equipment perspective, the telecom industry is looking at 26.2% CAGR to 2011, which is approximately 24 billion dollars. So, from a services aspect there is a huge business value proposition associated with CE. When we look at CE outside of the Metro, there is a tremendous amount of CapEx savings, 82% compared to the legacy.

If a provider has speed, multiple services over the entire network, enhanced customer service, increased ARPU, and low deployment costs, then convergence can be considered successful. The SP is able to increase revenue or profit for bit per type capability, while promoting and creating customer loyalty through an enhanced experience. Given these factors, Ethernet becomes the deployment option not only from a capability standpoint, but also from a financial one in its ability to: deploy multiple services; provide better NNI capabilities for interprovider-type of connections; simplify the network while simultaneously looking and identifying the interface of choice, DSLAM capability or switching router; and provide a possible solution for OpEx associated with mobile backhaul. What could be a more natural choice?