Virtualization of the online charging systems (OCS) function is a good starting point for operators that want to gain experience with virtualization before tackling larger, more complex network functions virtualization (NFV) projects. Not only is the OCS function a key component within a full IMS implementation, but with adequate performance and scalability metrics, virtualizing it will bring major benefits to CSPs.
Virtualized OCS is independent from underlying hardware; it is free to run on any COTS server, anywhere within the operator's organization. This implies that the OCS can be deployed wherever it makes the most sense in the virtualized infrastructure, and instances of the OCS can be replicated in different geographies rapidly; and the OCS can be upgraded and new features can be added faster, since development takes place in software only and is no longer constrained by hardware development cycles.
In the future, we believe that the policy and deep packet inspection components of the billing support systems (BSS) ecosystem will work in close alignment with the software-defined networking (SDN) controller. Policy rules, subscriber context, user transactions and behavior, application user trends, congestion events, device trends, etc., will be fed into the SDN controller in real time. This will help in creating self-managed network scenarios that will continuously adjust resources based on policies, quality of experience, efficiency and other parameters. (See BSS Is No BS and NFV Policy Wake-up Call.)
Billing and charging also need to interface with the self-service engine to provide the necessary usage/chargeback information to users, and must capture information regarding the provider of the service or event so that settlement can be accurately calculated and partners compensated for their contribution. Flexibility of pricing, complex bundling capability, embedded business intelligence to provide customized offers, etc., are critical features of a revenue-management solution set in an SDN/NFV world.
The main understanding with SDN and virtualization is that the intelligence of the network will now exist in a consolidated management layer (the SDN control plane). Product managers will be able to define triggers based on network events, chargeable events or billing status that can be combined in ways to provide customers with automated discounts and offers based on their use of the service. These can also be used to drive the network to provide greater enhancements or differentiation of service driven by those triggers.
In order to enable these types of services in an SDN environment, a modern-day billing and settlement platform must be deployed with the following traits:
Programmable, able to capture, and bill for any event: An SDN/NFV environment will require a billing solution that is also programmable and able to capture any type of chargeable event or trigger as soon as it is plugged into the control plane. Billing platforms also must support an unlimited array of charging/rating scenarios.
Respond in real time: When implemented, SDN will demand billing systems to respond in near real time to accommodate new types of billing events and to take policy-driven actions in response to the continual triggers from the control plane.
Be service-aware: In order to take advantage of the flexibility of SDN, the billing support system must be flexible in how it manages and bills for services. In other words, those operators who look to SDN/NFV will need a billing and compensation platform that is just as flexible, versatile and capable as the SDN architecture itself.
Provide compensation across various providers and channels: As previously discussed, the operators ecosystem will change even more rapidly with the dawn of SDN/NFV. Operators will be able to create new service models in partnership with any number of third-party vendors. Services will also be exchanged across various channels. With these new service mashups, as it were, operators will need a billing system that is able to accommodate multi-party compensation and settlement.
— Ari Banerjeee, Principal Analyst, Service Provider IT, Heavy Reading.