With many predicting 2017 as the year in which telco service cost per bit will exceed achievable revenues, operators must act quickly and effectively if they are to avoid large parts of their revenue streams becoming non-profitable.
With this in mind, Network Functions Virtualization (NFV)) has captured the imagination of the telecommunications industry. According to iHS (formally Infonetics), 35% of Communications Service Providers (CSPs) are planning to deploy NFV in 2015. Will they all succeed? I suspect not.
The fact is, there are many ways in which NFV could simply perpetuate the status quo. In deploying NFV CSPs may be deploying a ‘new legacy’. That is, creating software-defined silos where previously there were hardware-defined silos. These software silos can be just as inflexible, just as expensive and just as locked-in as their physical counterparts.
Simply moving to software – while truly powerful – only solves part of the challenge facing CSPs today. If the only thing you do is move from hardware to software, history is destined to repeat itself. This movement is bigger than just the adoption of new technology. It’s about the fundamental transformation of an industry, its culture and its approach to operations.
I’ve looked back at the last year or so at VMware’s growing list of CSP customers and identified three key common principles that I believe have made a significant contribution to our customers’ NFV projects’ successes.
Principle 1. Embrace the new supply chain.
- At TMForum Live! 2015 one speaker asked the question “Is there a win-win for operators and the Network Equipment Providers?” Ouch, that get’s right to the point. The reality is the supply chain must change if NFV is to deliver its full potential. Traditional vendors still have a huge amount to offer, but much of the core skills, innovation and operational know-how associated with NFV and cloud-based services exists among new market entrants. Vendors with whom the industry has not traditionally dealt. CSP project boards should ensure they engage with a healthy combination of key vendors and absorb multiple points of view before making key architectural and deployment decisions.
Principle 2. Adopt a ‘Platform First’ perspective
- Virtualizing an app is not the same as building a platform; a virtualized stack may have as much lock-in, rigidity and operational cost as a physical silo if the elements are not designed and built to enable true multi-vendor operation and support. To deploy NFV in a way that will deliver a positive business case, all VNFs must be accommodated on a single platform, like the teams at Vodafone Germany or Ooredoo in Kuwait have done. Only in this way do the economics and agility of cloud operations materialize. By owning and managing the platform, the CSP not only ensures the success of NFV, but also places themselves in a position of authority over the ecosystem of VNF vendors by generating open competition for the right to be chosen for the platform. A true best of breed option for the CSP. From the outset, building an independent platform capable of multi-vendor VNF support should be a key and separate requirement of any NFV project.
Principle 3. Recognise the need for Organizational Change
- NFV is more than a technology transformation. Deploying the technology without implementing organizational change is a major threat to the success of NFV. Why? Because it locks the CSP into the traditional technology, deployment and operational models that perpetuate the high operational costs and restrict the ability to innovate. It will take time to build the cloud DNA necessary to fully exploit the power of NFV but the journey must start in parallel with, not after, the technology transformation. Identifying new processes, roles and responsibilities for deploying, evolving and operating NFV must be a major focus of any NFV project.
Operators need to move quickly to embrace new supply chains, adopt a platform first approach and recognize and implement organizational change to take full advantage of NFV. Those that do will realize the cost benefits of the cloud model today, while building valuable cloud DNA to capitalize on the service innovation potential of tomorrow. Those that simply change technologies without changing the organization could be implementing ‘new legacy’.