Blueprint: 2016 and the Rise of NFV – Practicality Rules

by Martin Taylor, CTO, Metaswitch

With network function virtualization quickly moving into the mainstream and proliferation of related technology offerings on the rise, clarity of purpose and ease of use is more critical than ever. Winning solutions in 2016 will combine purpose-built technology with turn-key simplicity, making it easy for network operators to understand, adopt and scale NFV deployment system-wide.

Here are some of my predictions for 2016:

1. Pragmatic network operators in 2016 will progress the fastest; those who deploy proven VNFs that are not too demanding on cloud, SDN, orchestration or OSS/BSS integration will usefully move the virtualization needle in 2016. Leading solutions will:

Deliver high availability on vanilla cloud infrastructures without relying on a specially-engineered cloud infrastructure to deliver high availability.
Require only basic IP connectivity from the NFV network fabric, vs. requiring a high degree of programmability to create service function chains.
Have simple life-cycles and be able to deliver most of their value with little or no orchestration beyond initial deployment, vs. requiring sophisticated orchestration.
Have few and simple OSS / BSS touchpoints, rather than having complex configuration and management requirements and involving a lot of custom work to interface them to OSS and BSS.

2. VoLTE and CPE will be the two most active areas of the network for NFV-based buildouts in the coming year.

VoLTE is a service that requires a number of network functions to be deployed including IMS, SBC, TAS and SCC-AS, all of which are available in virtualized form.
Many services offered by network operators require the deployment of multiple items of CPE, e.g. Metro Ethernet access device, firewall, WAN accelerator, intrusion detection system, enterprise SBC – each of which is currently deployed today as a separate physical appliance. NFV offers the opportunity to virtualize all these functions and deploy them as software in a generic CPE device based on a server, or in a service provider’s cloud in a metro data center, thus removing the need to ship and install a multiplicity of physical appliances on the customer premises.

3. While 2016 will see NFV cloud and orchestration solutions mature, OSS/BSS will emerge as the biggest brake on NFV progress.

There are two issues here. First, integration with OSS / BSS is usually the long pole in the tent when it comes to deploying any new network function. There are numerous backend systems that a network function needs to talk to for provisioning, configuration, alarms, performance reporting, etc., and integrating with a network function at each of these touchpoints often requires custom software work. This issue does not go away just because a network function is virtualized.
Secondly, traditional OSS / BSS is not well suited to managing virtualized network functions because its view of the world is appliance-centric and it doesn’t know how to handle shifting populations of different kinds of virtual machines that together do the work of a physical appliance. OSS / BSS needs to evolve very substantially to cope with the realities of NFV, and this will take time.

About the Author

Martin Taylor is chief technical officer of Metaswitch Networks. He joined the company in 2004, and headed up product management prior to becoming CTO. Previous roles have included founding CTO at CopperCom, a pioneer in Voice over DSL, where he led the ATM Forum standards initiative in Loop Emulation; VP of Network Architecture at Madge Networks, where he led the company’s successful strategy in Token Ring switching; and business general manager at GEC-Marconi, where he introduced key innovations in Passive Optical Networking. Martin has a degree in Engineering from the University of Cambridge. In January 2014, Martin was recognized by Light Reading as one of the top five industry “movers and shakers” in Network Functions Virtualization.