IMS NFV from the Cloud

Infonetics Research hosted a webinar earlier last week entitled Migrating to IMS in the Cloud with NFV. That’s three of my favorite topics right there: IMS, cloud and NFV. IMS NFV can be consumed in two ways: build the cloud or use the cloud. That’s two options: IMS in the cloud or IMS from the cloud. There is an important distinction to be made between cloud technologies and cloud X-as-a-service solutions. The former requires building, the latter is about leveraging what’s already built.

Cloud and NFV and are not synonymous. NFV is about technology and network evolution. Cloud is about service delivery and business model transformation. NFV is employed to build a cloud. NFV and cloud technologies means CAPEX, fixed and variable OPEX and the same old operational model. Cloud service throws that model out the window. You could liken this to what Tom Nolle of CIMI Corporation has called this VNFaaS.

Returning to the focus of the webinar, Principal Analyst Diane Myers led the discussion and shared NFV adoption plans and vendors shared their views on challenges and key considerations. In an supporting piece of research, IMS in an NFV Environment: A Market Outlook, Infonetics asks an intriguing question. It’s the same one we’ve been raising given the cost and complexity of IMS and VoIP networks juxtaposed with the state of voice services market.

How do you justify spending capital on a new network for a service (in this case, voice) that represents declining revenue?

The reality: In most cases, you can’t! Service providers need to challenge the assumption that in order to deliver and monetize voice that they need to rebuild the network with IMS (with NFV or hardware), is the right approach. The “we’ve always done it this way” approach need not be the path forward, especially given that the need for change is now.

The Infonetics report also shows strong support and a desire to move quickly.

Service provider respondents overwhelmingly stated the desire to run IMS networks in NFV environment in two years as opposed to purpose-built hardware and dedicated servers they have today.

The reality is that NFV architecture is in flux, interfaces undefined and vendor gear immature. This 2016 is a start to a multi-year process and is nowhere near an end-date.

The main motivations?

The big driver for NFV is the ability to scale services up and down quickly and introduce new network services more efficiently and in a timely manner, which makes IMS a natural early hit for NFV.

The reality: The focus on voice should be on operational simplification and maximizing margins despite the headwinds and market dynamics. Manage voice for the best cash across your entire subscriber base vs. chasing some whiz-bang features that would get used by a small fraction of your users! How many new services or innovative features are there for voice? I think the main end-user innovations that remains is embedding and extending communications into different environments (e.g., home phone extended to tablets and smart phones, business VoIP in Google Apps, CRM systems, etc.). Yes, there are new services to address SMB and specific verticals and for some providers there is a net increase in revenues. However, the transformation that service providers most need is ease of management and a better, success-based business model.

Speaking of operational simplification, the representatives from Huawei and Metaswitch also admitted that NFV will make the integration and operations more challenging due to the many more moving parts in a NFV network (servers, hypervisor, many instances of VNF software, MANO layer and OSS). It’s the exact topic I explored in The New IP (see Virtual Machines, Real Complexity).

We are excited about this interest in fundamental shifts in technology change. For voice we think the service providers should re-examine who they are and what they need to own. Next gen voice should stretch beyond using cloud technologies and extend to using the cloud.

Don’t build an IMS cloud, use the one that’s already there.