knowledge centre

Telecoms network virtualisation trends and research

Larry Goldman Head of Telecoms Software

What are the challenges and successes of network virtualisation as it moves from proofs of concept to commercial deployment?

Video interview

Larry Goldman, Partner at Analysys Mason and Head of Telecoms Software and Networks, offers insights into the impact of software-defined networking (SDN), network function virtualisation (NFV) and cloud computing on OSS/BSS architecture and software market segmentation.

 

Listen to or download the podcast

 

Video transcript

I want to talk about the research we are doing in the area network virtualisation. 

This is an area of great interest and promise for network operators and a very signifcant issue for the vendor community.

It really is disruptive of the way that network operators have deployed their networks up until now. 

It's an area that's been getting a great deal of interest for the last several years and we've done a great deal of research until now but it's becoming more and more central to the approach we are taking in our research and that's really reflective of the fact that it has become more and more central to the network deployment strategies of operators all over the world.

So as we approach this research, the core of what we are doing and what really focuses most on this, is the programme we call Software-Controlled Networking.

In this area we forecast the operator spend for cloud, for network function virtualisation (NFV), for software-defined networks (SDN) and we have done this for quite a while and we continue to break this down and get more detailed in how we do this. 

We also lay out the overall strategies and and how they are developing. We explain what different operators are doing around the world. We explain how those strategies differ from one another and so forth. 

It is certainly a world where to teir 1 operators are way out in front. We are seeing examples of what they are doing, and explaining to other operators what that means and certainly for the vendor community what it means. 

What we really see going on is that there is a whole new value chain of the way networking value is being deployed by network operators. 

It's very much going away from the idea of putting in a discreet boxes in networks and much more of building a network platform, a virtualised network platform, that runs on more industry-standard IT hardware and the network functions that we use to embody in boxes are now virtualised in software that runs these standard platforms. 

It's actually a fairly easy concept to talk about at a high level and very difficult one to actually deploy and make work in the real world of network operations. So that is what our research is really grappling with.

Beyond this strategic view that we are laying out of the overall approaches in Software-Controlled Networking, we also have serveral programmes that take this into specific areas of detail.

One of those areas is Next-Generation Wireless Networks, where we are really looking at how wireless networks get built in this new virtualised world. What characteristics do they have? How are the technologies changing? What does the value chain really look? 

Another program we have that is very much wrapped into this "how do you deploy capabilities?", is our Service Delivery Platforms research program. It is a programme we have had for quite a while and we have seen that this area has changed a lot in the last few years as many of the things that have been built have been put into a virtualised environment.

So those are areas we've been working, specifically around the network.

We believe that one of the significant changes that is happening as we apply the concept of virtualisation is that operators need a new idea of how they operate a network. We call it a next-generation operational framework.

The core issue here is that  there are a lot of things that can work in today's environment, with discrete boxes and so forth, people can walk up to those boxes and deal with them. It is a very manual process and it works for operators. 

Part of the promise and problem of virtualisation is that that manual process doesn't work anymore, so operators really need a fully-automated view of how they are going to operate their networks.

That calls for this next-generation operational framework. Within that framework a lot of different problems need to be solved including: the network itself; the management of the network; the handling of problems in the network; and the instantiating of services in the network.

All of those things require this next-generation operational framework and that framework we are mapping out in our Network Orchestration research programme. 

Our research is mapping that out and explaining how that (framework) applies in things that we've already seen happening and how we think that that's going to continue evolve and develop. 

Again, it is part of this new value chain for the way operators will develop capabilities and what that means for vendors in terms of how they will have to sell their product capabilities into this new virtualised network and next-generation operational framework.

So then we we have several programs that continue to cover the traditional areas of OSS including Service Assurance and Service Fulfilment research programmes. These programmes continue to deal with the reality of where we are.

We have real networks today built with those traditional network elements and people are migrating from that existing world and making it work. It is how their networks operate today. That's how they generate revenue today. And now they are now migrating to this new next-generation operational framework virtualised network. So in our Service Assurance and Service Fulfilment programmes, we are covering how operators make those migrations and what does that mean for vendors.

So we have a group of programs that are strategically organised under the umbrella of Software-Controlled Networks and then we have programmes that deal with specific aspects of different elements of dealing with the vritualised network.  

It is an area that has enormous promise. It is very disruptive for the operator and vendor communty. We think our research is really helpful to all parties involved to understand not only where they are trying to get to but how to they get there from where we are today.