Introduction

We're eight years into the network functions virtualization (NFV) journey, starting from when leading global carriers assembled their visions into the seminal paper that defined the foundations of NFV. It's taken the industry a while to come to grips with all the machinations needed to make NFV operational and performant, and we've collectively learned a lot. We've evolved NFV from simple virtualization on a single machine to cloud-native implementations. That has proven especially important in 2020 as the COVID-19 pandemic spotlighted the critical role that the telecommunications sector plays in keeping businesses and communities connected. As part of their evolution, network operators are looking to 5G and the edge as the next wave of innovation.

5G is a rallying cry that has commandeered a large amount of attention and capital investment from carriers, and rightly so. 5G represents a serious upgrade to networking services for consumers, enterprises, and the public sector. It promises massively improved bandwidth, unprecedented latency reduction, superior reliability as well as tremendous flexibility. Also, by supporting differentiated services, 5G provides opportunities for new business models for communication service providers (CSPs) not possible with 4G. To enable these differentiated services, CSPs will need to evolve NFV to encompass cloud-native virtualization and embrace innovation in orchestration platforms.

In this research brief, we'll look at the evolution of NFV into the world of clouds and explain why successful 5G rollouts require dynamic network function (NF) placement. We will explore the topic of NF placement in detail and provide critical considerations for the move towards 5G.