At TM Forum Live! 2017 in Nice in May, Dr. Vikram Saksena, Chief Solutions Architect – Office of the CTO, NetScout, will deliver a case study with Swisscom on service assurance in the era of virtualization. Here he looks at the changing role of service assurance in virtualized networks.
As we continue to move towards 5G, carriers are under more pressure than ever to deliver a consistent Quality of Experience (QoE) for a user base now heavily reliant on data services. This is having a direct impact on the way they look to approach new service rollout, and also service assurance. Carriers are increasingly making plans to virtualize new service offerings and phase out physical network hardware and make way for virtualized alternatives that are designed to simplify operations and improve network functions previously performed by proprietary hardware. But they are also realizing that making this a reality will not be straightforward.
As is the case with any significant industry shift, carriers are currently faced with a series of challenges associated with the move to network functions virtualisation (NFV), most notably in terms of service assurance.
The role of NFV in modern mobile networks
NFV promises to virtualize legacy infrastructure and create an environment far more agile and adaptable than today’s infrastructure based on physical systems. The challenge carrier’s face, and part of the reason NFV has become so important, is that legacy networks are notoriously difficult to modify or upgrade. With carriers still looking to recoup the costs of their significant investment into LTE and IMS and with subscriber data demands higher than ever before, it’s hardly surprising that existing physical infrastructure is being stretched to try and accommodate this need.
The fact Over-the-top (OTT) applications are software-based and therefore capable of running on top of the mobile network means they can be rolled out almost immediately, creating a steady stream of data-intensive applications capable of stripping network capacity and instigating new peaks in demand. Carriers need a more adaptable approach to ensure their mobile networks can live up to subscriber demand, putting service assurance at the forefront once more.
The importance of service assurance
There has always been a heavy reliance on network monitoring and troubleshooting tools to support legacy networks and quickly identify where subscribers are encountering issues. However, monitoring and optimizing performance in a virtualized environment, particularly in response to ever-increasing data traffic demands, is more challenging. NFV gives carriers a far more flexible approach to new service and network functions rollout. However, it also creates a huge challenge to the teams responsible for network/service/customer assurance.
Carriers not only need a horizontal view of all network activity (from the RAN, to the Core, to the cloud), but also a vertical view, covering the full network stack, for signalling and the user plane. To avoid the risk of service disruptions, increased churn rates, and damaged reputations when moving to software-based network architecture, it’s essential to have access to powerful tools capable of giving a real-time, end-to-end view of subscriber activity across both the legacy physical and new virtual networks.
Getting it wrong will have devastating implications for subscribers and could potentially undermine NFV as a solution to the very problems it’s expected to resolve. Fortunately, the introduction of NFV has paved the way for easier handling of service assurance.
Improving QoE in a virtualized network
Being able to deploy network functions as software rather than hardware through NFV means the network can be easily scaled in response to subscriber demand. Given the costs associated with either scaling in/out of the network, it is imperative that network re-sizing is done in the most effective manner. It is heavily dependent, therefore, on having the right information in the right place.
It’s for this reason that service assurance can no longer be taken in isolation, particularly not for a virtualized network. Being able to ensure that a carrier’s network does not falter during peaks in traffic, and to guarantee subscribers do not face issues like dropped calls or a loss of data packets, relies on having full visibility of what is going on across the board.
Automating the virtualized network
This evolved approach to service assurance mixed with big data analytics has an important role to play for network functions virtualization to become mainstream. Network data can be used to inform and adapt the day-to-day running of the network as part of a feedback loop for the orchestration layer. This will improve the QoE and reduce subscriber churn. Plus, it will have a dramatic impact on network management across the board.
By capitalizing on tools that provide a real-time view of the entire network environment, spanning legacy and virtualized hardware, carriers will have access to a broad range of meaningful network data and subscriber information. This data will let them predict peaks in traffic and put processes in place to ensure that high-value subscribers consistently get a quality experience.