Ahead of his presentation at TM Forum Live! Dr Lester Thomas, Chief Systems Architect at Vodafone Group, looks at the critical role APIs and platforms will take in driving new revenue streams from SDN and NFV.
Many telcos are investing heavily in network virtualization technologies, even though there is not yet a clear vision of where the business benefits lie. There is a belief that there will be future new revenue and cost-efficiency opportunities enabled by virtualizing our networks. This belief comes in-part by looking at the experience of the hyper-scale web companies: Virtualization of infrastructure services has transformed the business models and architecture underpinning every online service we use. Virtualization has enabled unprecedented levels of innovation, agility, scalability and end-user experience. It has also reduced the barrier-to-entry for new start-up companies which have been the revenue growth engine for the Infrastructure and Platform-as-a-Service providers.
Building network capabilities for services that have not yet been invented
You can imagine that adding Virtualized Network-as-a-Service to the existing compute and storage services will drive a new round of innovation and growth – particularly when added to other growth engines like Internet of Things or the extension of telco services into other industry verticals. In such an emerging technology and business area, how do we define our future business operating model and technology architecture? We need to build the next generation of our telco industry based on these virtualization technologies, but there are many challenges: For example, where should we standardize to build an industry ecosystem and where should we drive competition and innovation?
There are many industry standards bodies working on the technology-layer standards for Network Function Virtualisation and Software Defined Networking. We may be able to achieve some cost efficiencies by adopting these standards in our existing telco businesses. Simply integrating these SDN and NFV into existing OSS and BSS systems, however, will not unlock the potential new revenue streams that virtualized networks offer. We need to re-define our BSS/OSS architecture to fully exploit network virtualization and to drive new revenue streams from products and services that don’t exist today. Some of these will be for new end-to-end services that only become feasible with virtualized networks – how do we build underlying network capabilities for end-to-end services that have not yet been invented?
To be successful, we also need some standardization of the BSS and OSS layers to enable an industry ecosystem. We know that a smart virtualized network will enable us to deliver the best possible Quality of Service (QoS) at the lowest cost. There are alternative over-the-top solutions like Virtual Private Networks (VPNs) that can deliver some of the networking requirements and so, to compete as an industry, we have to build telco standards for network virtualization.
Interoperability and innovation?
A key question arises: How do we standardize and enable interoperability, whilst enabling innovation (both in terms of technology implementation and business models)? One lesson from the hyper-scale internet companies is that APIs and platforms allow the creation of a standards-based ecosystem, whilst enabling innovation in the implementation. When you consume the services of one of the Infrastructure- or Platform-as-a-Service companies, you integrate through standards-based APIs, yet have no knowledge of the underlying systems or processes that deliver these services. The platform service providers are also constantly innovating in how they deliver these services, using big data and now increasingly machine learning to constantly optimize the customer experience and the cost-efficiency. Could we apply the same approaches to a standardization of our BSS and OSS systems to fully exploit the promise of SDN and NFV?
A critical advantage of platforms and APIs is that they provide an evolutionary path from the current telco operating model to the future Network-as-a-Service model. One aspect often overlooked is that we will be running hybrid networks for many years. Even in the target-state, there will always be physical components in our end-to-end services. APIs provide a way of abstracting away from the implementation of a service and so enable interoperability of legacy physical services with next-generation virtualized services. A continuous evolution roadmap on which to build our future telco revenues is also possible.
How do we define these layers of platforms and APIs that will form the blueprint of our future architecture? Join me at TM Forum Live! where I will discuss these challenges and propose a new API and platform-centric approach to realise the true business benefits of SDN and NFV.