Making sense of convergence
Convergence is a powerful word. For mobile operators, those 11 letters hold great promise. They imply more efficient networks that address today's services and prepare for the arrival of new ones, such as LTE.
In theory, any action related to convergence should make things better. Better end-user experiences with new types of services and capabilities. Better response times, cost savings and revenue streams for operators.
But, in implementation, the concept isn't as clear. Convergence can mean a lot of things in different parts of the network. Just like you can't fit a round peg into a square hole, you can't apply one network convergence strategy to every area of your network.
Make the user plane simpler
First, there is the user plane. In the mobile backhaul space, we've already seen convergence at both the physical and virtual, or service, layers. Plus, the benefits of converging traffic across generations of radio service (2G, 3G, 4G) is well known.
But, we're also starting to see blurred boundaries between transport-like devices and switches, routers and other data plane devices. In addition to incorporating Ethernet switching into optical transport systems, switches, routers and cell site devices, service providers are starting to explore the use of optical technologies like OTN. This gives you two distinct advantages: more resiliency and simplicity.
Reduce costs in the control and management planes
In addition to user plane convergence, there are also opportunities to converge technologies and network elements through the control plane and the management plane.
Some of the most important benefits of any network convergence are operational cost reduction and simplicity. In this sense control plane and management plane convergence are among the most important—even more so as operators pursue best-in-class technology and need to make it seamlessly fit into a larger ecosystem.
Backhaul and packet core belong together
Evolution in the packet core space converges various generations of 3G technology with 4G or pre-4G standards such as WiMAX or LTE. In some cases WiMAX operators will move to LTE themselves, providing another rationale for a converged packet core.
A packet core that serves all of these personalities together makes for a seamless evolution. It enables operators to have different generations of traffic coexist with one another. You also have the option to migrate to newer technologies at your own pace.
Making convergence work
This then begs the question of how we converge mobile backhaul and packet core to take advantage of the tight link between these parts of the network.
One goal is to converge synchronization and resiliency mechanisms, failovers and so forth all in the same manner. This is a much more efficient long-term scenario for service providers.
Plus, this distributes intelligence toward the end user. When you move packet core capabilities into the mobile backhaul, you gain long-term advantages of LTE related to the RAN and standardization of interfaces.
Finally, by integrating deep packet inspection (DPI) with packet core, you can tie it to capabilities in the backhaul and map classes of service to Quality of Service. Then, you can base your resiliency decisions on visibility into the flows and enforce traffic prioritization in your network. This built-in DPI makes convergence work for end users and operators. Consumers get more choices for services that are useful or entertaining to them. Operators have more ways to generate revenue and maintain a loyal customer base.
The link between mobile backhaul and packet core gets tighter over time as networks transform to IP. Convergence is the way to make the network better for end users and operators.
And that makes sense.
