How close are we to 5G edge cloud?

New, enhanced, and useful customer experiences are vital to the successful adoption and monetization of new 5G services. 

As millions more devices connect to their networks, telecommunications service providers are migrating from hardware-based network appliances to virtualized infrastructure to enable them to rapidly and economically scale to meet ever increasing demands from customers. 

To deliver reliable 5G services, one way operators can improve application performance and reduce latency is by extending telco cloud infrastructure from their network core to the edge: closer to customers, devices, and data sources. 

5G Edge Cloud Survey

In January 2020, Heavy Reading undertook a comprehensive global survey of operators’ 5G strategies. Red Hat sponsored the research into operators’ edge cloud deployments. 

The survey revealed four main motivations for deploying edge cloud infrastructure: 

• Ensure application performance

• Delivery of vertical industry services

• Reduce bandwidth use and lower operating costs

• Offer differentiated communications services

In a webinar discussing the survey, Azhar Sayeed, chief architect, service provider at Red Hat, said, “A few years ago, the focus was on reduced bandwidth and lower operating costs. These latest findings show that application performance is now the primary driver for edge cloud deployment, and you can actually see a shift in the trend for vertical applications support.”

How close is 5G Edge Cloud?

5G is expected to enable unprecedented levels of data capture, processing and storage at the edge - allowing operators to deliver new services such as digital healthcare, smart traffic management, connected and autonomous vehicles, machine sensor processing and pre-emptive maintenance, and real-time analytics. 

While more than 80% of service providers told Heavy Reading that they had started deploying network edge cloud infrastructure, just 19% described theirs as "mature." However, most respondents expected to have hundreds of edge cloud locations by 2023, with 34% indicating they would support thousands. 

Source: Heavy Reading 5G Network & Service Strategies, 2020 Operator Survey

During the webinar, Gabriel Brown, Heavy Reading’s Principal Analyst, said, “If you include vRAN, then you will have a much larger number of sites.” Red Hat’s Sayeed further quantified these findings, adding that the US has approximately 220,000 cell sites and, even with aggregation, one operator alone would have 10,000 to 20,000 vRAN edge locations.

Radio Access Networks (RANs) at the Edge

Heavy Reading also asked operators how important OpenRAN and 5G will be to their networks. While Brown believes that classic RAN is currently dominant in 5G, he noted that some operators believe that vRAN will be an anchor tenant in the far edge. 

Sayeed agreed with this, stating, “If you look at evolution of new services for 5G and particularly the ultra-reliable low-latency [communication], the requirement can be only met when you have split RAN architecture, the cost justification is far better using a virtualized or cloud-native blueprint than your typical hardware.” He added, ”vRAN has almost become a necessity.”

Which Platforms are Operators using for 5G Core? 

Service providers are deploying edge clouds for multiple reasons, requiring them to support different types of workloads, running in virtual machines (VMs), containers, or both. Heavy Reading found that the largest group of operators (42%) actually plan to run their 5G Core as containerized applications in VMs. Brown commented, “Quite a number of the operators I talk to would be looking to use VMs and containers and VNFs and CNFs, all in the mix. One of the primary reasons for that is that their virtual infrastructure is now mature. It’s reliable; it’s resilient; it’s backed up with quite an effective operating model that’s taken them a good few years to get in place and rely on.” 

Source: Heavy Reading 5G Network & Service Strategies, 2020 Operator Survey

Supporting Operators’ Transformation from CSP to DSP

Red Hat’s open source cloud platforms lead the way in helping digital service providers transform their networks to cloud delivery and extend them to edge locations. Red Hat OpenStack Platform has become the open source standard cloud platform for network function virtualization (NFV), and Red Hat OpenShift leads commercial Kubernetes distributions, ready for cloud-native network functions (CNFs).

As an example, Turkcell, the first operator in Turkey to offer live TV over 5G, selected Red Hat OpenStack Platform as the foundation of its Unified Telco Cloud NFV infrastructure platform and reduced the time taken to launch new services by 66%. Additionally, Turkcell can scale capability in proportion to increased traffic, and readily onboard virtual network functions (VNFs) from the vendors it chooses. 

Similarly, Etisalat Group selected Red Hat OpenStack Platform for its SAHAB NFV telco cloud to support the delivery of digital services at scale across multiple sites. Anticipating the next phase of NFV and to accommodate CNFs, particularly for 5G, Etisalat is already experimenting with Red Hat’s OpenShift platform as an extension of its Red Hat-based NFV infrastructure.

Keeping the Edge Open

As operators transform into digital service providers that are delivering an array of vertical industry services, Red Hat is working in open source communities to simplify their transition to 5G and edge clouds with technologies like KubeVirt in its OpenShift virtualization.

Red Hat believes that keeping the edge open offers community-powered innovation, consistency of management, automation and orchestration, enabling service providers to launch new services and efficiently manage millions of devices at the network edge. As they navigate the transition to 5G and edge, service providers can rely on Red Hat’s open hybrid cloud technology to manage diverse workloads and deliver innovative, differentiated services, bringing them closer to consumers than ever before.