Unleashing the Potential of NFV

Network Function Virtualization (NFV) held great promise. NFV is the next generation of network architecture that replaces software running on dedicated hardware with virtual appliances running on commodity hardware. NFV meant being able to deploy software functions on any hardware at any time, with high interoperability so any functions could be switched and mixed between vendors. But the promise has not been fulfilled.

[[252878]]

Traditional telecom networks are a collection of dedicated hardware - each box, each site needs to be installed and maintained individually. This is the traditional way to build a network - large capital expenditures, long planning cycles, and often end up with vendor lock-in and uncompetitive pricing due to the complexity of interconnection. Upgrades and replacements are also resource-intensive in terms of integration and debugging.

Therefore, it makes business sense for the network world to move towards an NFV model: providing a means of fast and flexible installation, as well as helping to deliver predictable operating expenses and ultimately enabling communications service providers (CSPs) to benefit from reduced overall costs.

Software Interoperability

To truly achieve this, software elements need to be fully interoperable to enable vendor independence and competitive pricing. The resulting network scales quickly, is flexible, and benefits from dynamic resource allocation. This is what NFV should be - a wide range of interchangeable trusted virtual network functions (VNFs) that can be easily and cost-effectively deployed.

The reality is that the lack of information and insight means CSPs are being locked into full-stack virtualization solutions from a limited number of vendors. NFV interoperability is nowhere near where it should be, and attempts to address this are ongoing. NFV MANO (Management and Orchestration) is an architectural framework that facilitates the deployment and connectivity of services as they migrate from dedicated physical equipment to virtual solutions. However, ETSI descriptors are only partial, allowing vendors to maintain some proprietary elements. This means there are gaps in the information used for virtual function operations.

Similarly, the ONAP project aims to create a comprehensive platform for real-time, policy-driven orchestration and automation of physical and virtual network functions. Its goal is to enable software, network, IT and cloud providers and developers to quickly automate new services and support full lifecycle management. However, the influence of major vendors is very strong, and they still want to keep their proprietary elements opaque.

This meant that sourcing from multiple vendors was problematic. Compatibility was not a given, and integration that should have taken minutes would take days or even months, as the onboarding of each new virtual function was unnecessarily complicated by this partial interoperability. And vendors kept updating software versions, and the painful process had to be repeated from scratch every time a new version was released.

We also need to remember that the shift to virtual networks is not a revolution, but an evolution. Sure, there will be new NFV-only networks, but there are billions of physical network elements that still have a long lifespan. NFV can help the network change with market demands by placing virtual functions where flexible functions are needed (functions that can be turned on or off as needed). This hybrid network that combines physical equipment with virtual functions will bring greater operational efficiency and speed to market for new services.

Today, the reality is that customers who purchase VNFs may spend months trying to integrate those functions, so NFV adoption and growth has been hampered. NFV should unleash the power of innovation, which will become increasingly important as 5G networks and virtual functions push computing power to the edge. For applications such as augmented and virtual reality, powerful edge computing capabilities are needed - without NFV, 5G will not be able to deliver even half of what it promises.

NFV Technology

What the industry needs is an authoritative source of information about NFV solutions, and a consistent, descriptive, searchable taxonomy. There also needs to be a service that tracks and manages certification and standards compliance, so compatibility can be determined before implementation. Without a clear registry, NFV will not have the transformative impact it has on our networks.

By introducing a single reference point to compare VNF policies and software versions, network operations will be improved, network security will be enhanced, and broken, malicious or outdated VNFs will be identified. Network operators will be able to better understand which functions will operate in their specific environment and can quickly integrate to promote the benefits of NFV.

Confusion and frustration have delayed service provider deployment of NFV. The complexity of connecting traditional physical components and launching a virtual approach to delivering, ordering, assuring, maintaining and billing services has not been clarified. Lack of information, lack of automation and excessive vendor lock-in limit the value proposition of NFV to communication service providers.

Gartner says the promise of NFV will drive "robust revenues from digital services and enhancements." If that happens in the future, it's time for the industry to standardize the information and tools needed to create, manage, and effectively operate virtual and physical network assets and functions. It's time to centralize expertise and streamline NFV environments. The result should be reduced complexity, good scalability, clean automation of business processes, and beneficial optimization of assets. Only then will NFV be unstoppable, leading to greater operational efficiency and improved customer satisfaction.

Original link: https://www.rcrwireless.com/20181207/reality-check/unleashing-the-potential-of-nfv