NFV and SDN are promising transformations that will improve customer satisfaction/retention while reducing OPEX. Much has been said about NFV's ability to accelerate service activation, and how SDN will help to assure service performance and increase service monetization. Fortunately for innovators, these are disruptive transformations, but many are still wondering how to transition their existing networks and OSS, and which technology evolution they should prioritize.
As networks are increasingly virtualized and become more dynamic, our industry is being forced to re-invent itself and streamline existing network management practices, all in the face of decreasing ARPU and increased customer acquisition costs. This brings an interesting challenge in that communications service providers (CSPs) must dramatically alter how they manage and monitor their networks while still offering the best customer experience and without losing customers in the migration process.
Therefore, every innovative CSP is seeking answers to the following fundamental questions:
- How can I assure network service performance and quality over my virtualized and dynamic networks?
- How can I ensure that NFV and SDN migrations will not come at the expense of my customer base?
It has always been striking to me that service assurance and performance management are afterthoughts. Product management teams shape innovative offers and network engineering teams design network service prototypes, but teams are only tasked with building the tools needed to properly manage those services once they've been launched. This seems backwards to me, especially as service quality is often a challenge for any service's first customers.
Several NFV/SDN management use cases are showing us that we are entering a new area where service performance assurance solutions are required at the service launch:
- Customers will only migrate to NFV/SDN based network services if the operator can demonstrate that service levels will be met. This requires that CSPs offer service level reporting through their existing customer portals.
- An NFV orchestrator will only be able to increase the capacity of virtual network functions (VNFs) if the underlying cloud infrastructure can handle the increased load.
- Network operations and engineering teams now need to manage the performance of distributed and multi-vendor NFV infrastructure in data centers, points of presence (PoPs) and on customers' premises, alongside their existing network infrastructure performance.
- To make the best decisions, the SDN controller will need visibility into the end-to-end network and service performance.
In that context, end-to-end and cross-domain network performance platforms are becoming mission-critical solutions and a key success criteria for NFV and SDN based services.
If you are interested in these topics, I recommend reading another article in which I explain why the ability to monitor the service delivery chain is a key capability for NFV/SDN success: Service Performance Assurance Enables Operators to Realize the Benefits of Their SDN and NFV Transformations. I also recommend listening to a webinar with Sergio Zveibil, which describes the advantages that mobile operators will gain by adopting a true end-to-end network performance monitoring strategy.
As usual, do not hesitate to contact me or connect with me on LinkedIn if you would like to discuss service assurance topics further. And, be sure to check back for a future blog post about the performance management maturity model, in which I will outline the necessary steps to ensure virtualized and dynamic networks maintain high performance.