July 22, 2022
Building the Converged Core for the Future
Building the Converged Core for the Future
Delivering on the 5G promises
In a previous article, we discussed the technologies enabling MNOs to think differently about the future of telecom networks because the market is changing. Building on this, we’re going to take a more focused look at the network core.
The fundamental promise of 5G is that different people and organizations have different connectivity requirements. From synchronized robots to systems for remote control of construction vehicles – some use cases need high capacity and high speed; some need low latency, and some need to have a lot of different connections in a small area. But achieving this connectivity ideal isn't possible with the traditional network core.
The new paradigm for a mobile packet core is not simply scaling the core in order of magnitude beyond the previous generation – it is a whole new ball game.
Scaling in the cloud
The technological backbone of the Packet Core of the future is converged, cloud-native, ultra-flexible, containerized, and infinitely scalable – giving operators a leap forward in their ability to provide new and innovative services.
But the technological stack is just a part of the story. The real excitement comes when operators use these cutting-edge capabilities to create meaningful, tangible monetization opportunities and innovate to create new services.
One of the key benefits is improved network reliability and disaster recovery (DR) capabilities enabled by separating the control and user planes within the network (CUPS). With the individualized control capabilities of these planes, there are multiple opportunities to win. For example, CUPS enables Multi-access Edge Computing (MEC) for network functions like internet traffic offload deployed at the far edge of the network, reducing overall network load and improving network performance for users.
A Converged Core can also leverage the flexibility of CUPS to enable Multi-Tier Resiliency for both the control plane and the user plane. Geo-redundancy and disaster recovery are achievable in each instance with a mixture of On-Prem Cloud and/or Public Cloud instances deployed anywhere. This flexibility creates peace of mind for the wireless industry that is somewhat unique. The hyper-scalers delivering access to the cloud for telcos have demonstrated their mettle in live packet core deployments worldwide, enabling everything from edge computing to remote control construction vehicles and smart factory automation.
Low-risk use cases to create market traction
The benefits of cloud-native Packet Core that we've already covered here look good on paper, but what are the practical implications for MNOs trying to balance the books and make systematic improvements across the network?
3GPP specifications for Packet Core network architecture emphasize modularity, reusability, and the self-containment of network functions as key design considerations. But to make this a reality, operators need to leverage multi-vendor compliance to reduce CAPEX/OPEX and avoid vendor lock-in for best-of-breed solutions. Through service-based architectures (SBAs) and open interfaces, operators can utilize the technologies and innovation strengths of numerous OEMs related to various network functions in Converged Core.
This multi-vendor approach also ensures that the baby doesn't get thrown out with the bathwater through interoperability with legacy core and radio access network (RAN) infrastructures, ensuring operators can manage costs as they make significant decisions about their future network. To that end, a modern Packet Core should be able to integrate with an existing legacy environment and natively provide support for 4G and 5G services, either Stand Alone (SA) or Non-Stand Alone (NSA).
To illustrate the value of a modularized approach to Packet Core deployments, let's look at the NEC user plane function (UPF). It's a fully containerized microservices-based network function that provides 640 Gbps of peak throughput per server. But let's get more specific with a low-risk use case by looking at a private cloud 5G deployment, which entails custom scalability aspects of all network functions. Operators can deploy UPF on the edge along with RAN workloads, creating a low-footprint Packet Core solution.
Flexibility creates long-term value
So do we get all the benefits we talked about up top with this solution? In short, yes; disaggregated SBA and micro-services-based network functions are scalable on-demand to cater to different use cases, from centralized network deployments to MEC deployments (coupling UPF and RAN) on the edge. They also support CUPS Architecture, and we've seen a ~20% increase in throughput peak performance per server on the UPF compared to traditional core networks.
A fully cloud-native and platform-agnostic approach supports continuous integration and deployment (CI/CD), service automation and orchestration, and life-cycle management. But it also needs to be multi-vendor, 3GPP with advanced feature readiness, and multi-technology compliant. The complete converged core also supports 4G/5G deployments and integrates with incumbents to support 2G/3G scenarios.
One final consideration that makes this a compelling Packet Core solution is its low carbon footprint, with a ~50% reduction in energy consumption. MNOs have considerable value at stake to improve their ESG (environmental, social, and governance) performance to comply with tightening ecological regulation and internal sustainability targets while reducing costs and enhancing brand image. NEC takes the sustainability of supply chains very seriously (platinum graded by EcoVadis), and we know it's a vision for a greener, brighter world shared by our partners and customers.
Beyond the user plane
Innovation in the Converged Core is critical because what's unique about 5G is that this is the first 'G' that is not equally relevant to enterprise use cases as it is to consumer use cases. From 2G through to 3G and 4G, as consumers, we could see that we were getting content delivered faster at better quality. We could watch videos in high definition and download media files faster. And of course, these same improvements are there for consumers with 5G.
But for the enterprise market, 5G is the first generation of mobile networks with sub-second latency, enabling some companies to revolutionize their business processes completely. Robotics, drones, and autonomous vehicles all need millisecond response from the network, and 5G is critical to achieving that. 5G provides different connectivity, which requires network slicing, which, in turn, requires a Converged Core. In this way, operators can create those connectivity products on 5G and – critically – on 4G given most networks are not pure-play 5G. So when we talk about latency, reliability, and throughput, we need to understand why improvements are necessary to Converged Core – which is essential to the entire network. And that's making these use cases a reality.
We know that for operators to consider a multi-vendor approach to deploying innovative solutions on their Converged Core, they need smooth integration into the incumbent system and interoperability with RAN and the 4G Evolved Packet Core (EPC). And that's what NEC Open Networks' Converged Core does — providing a cloud-native converged 4G and 5G solution.
We have a strong track record of providing operators with access to best-of-breed solutions on-premise, in the operator network, on AWS, and in hybrid mode to create the networks of the future today. Our teams have already deployed the NEC Converged Core commercially and at scale with Rakuten Mobile and NTT DoCoMo, as well as in two dozen private networks. And we're just getting started as we bring our suite of solutions for 5G to the rest of the world.