Last year, Red Hat shared news of the upgraded Singapore lab with Intel and Hewlett Packard Enterprise (HPE) designed to accelerate the adoption of 5G and edge computing. Since then, we have seen great momentum and interest from our telco customers and partners in the region to use this facility to test and validate joint solutions with Intel and HPE. For instance, one of our network equipment provider (NEP) partners used the lab to test a mobile core cloud-native function (CNF), which has now been put into production for a telco service provider’s network in Australia, helping them reduce risk and address any issues before deployment. Another NEP used the Red Hat Asia Pacific (APAC) 5G and Edge Technology Lab to test a blueprint for its 5G Core in preparation of a 5G SA core deployment project to help verify interoperability, performance and reliability. We foresee a rising interest in these activities with telco customers and partners as the demand for 5G SA core implementations continues to increase and open radio access network (RAN) and virtualized RAN (vRAN) projects materialize with pioneering telcos in APAC.
Through our work of bringing together NEPs, independent software vendors (ISVs) and systems integrators, Red Hat, Intel, HPE and our combined partner ecosystems have gained in-depth insights about modern deployment needs for service providers in APAC, which is further valuable for service providers in APAC to tackle the complexity of modern networks. After the initial wave of 5G non standalone (NSA) that re-used the existing 4G infrastructure to quickly launch 5G, service providers around the world have started to deploy 5G standalone (SA) core networks. Only 5G New Radio (NR) combined with 5G SA core enables advanced 5G use cases such as network slicing as well as some of the Industry 4.0 use cases, like AI-enabled visual inspection of factory output. Service providers are looking at these use cases to monetize their 5G investment.
From a technology point of view, 5G SA core is based on cloud-native Service Based Architecture (SBA) and consists of cloud-native network functions (CNFs) made of microservices hosted on a container platform. Many service providers such as Telefonica and Verizon have opted to use Red Hat OpenShift as a vendor neutral container platform because they source CNFs from multiple vendors and find value in standardizing on a single container platform, sometimes even across network and IT. Red Hat OpenShift is supported by an extensive ecosystem of validated CNFs and bare metal containers, which are well suited for data plane CNFs.
The Red Hat APAC 5G and Edge Technology Lab runs in parallel to on-going validation efforts between network equipment providers (NEP) and Red Hat, such as our ongoing collaboration with Ericsson, as well as the cloud-based onboarding service and testbed for network functions set up in North America by Red Hat with the support of Intel.
Introducing new offerings from Intel and HPE
The laboratory now offers HPE ProLiant DL380 Gen10 Plus x86_64 servers optimized for running CNFs with Intel technologies such as 3rd Gen Intel® Xeon® Scalable Processors, Intel® Ethernet 800 Series network adapters, and Intel® Optane™ PMem. The servers can be configured in standard Red Hat OpenShift clusters with a separate control plane and bare metal worker nodes entirely dedicated to hosting workloads. The servers can also be configured in smaller Red Hat OpenShift clusters to simulate edge configurations, including:
- Compact 3-node Red Hat OpenShift clusters wherein each server hosts both the control plane and workloads. Such a cluster can also provide built-in highly-available storage for Permanent Volumes (PVs) using Red Hat OpenShift Data Foundation.
- Single node OpenShift clusters wherein the control plane and workloads are hosted on a single physical server to provide feature parity on a minimal footprint.
All configurations provide Container as a Service (CaaS) and can also provide Infrastructure as a Service (Iaas) with virtual machines (VMs). Bare metal Red Hat OpenShift clusters can host VMs side by side with containers thanks to OpenShift Virtualization. To support the move to programmable networks, the laboratory also features a 100 Gpbs Intel Tofino based P4-programmable Ethernet switch. The laboratory facilities are accessible over VPN for easy access by the partner and can be extended into hybrid cloud configurations with site-to-site VPNs.
In addition, HPE and Red Hat have developed Telco Blueprints in order to accelerate the deployment of telco clouds and minimize the integration risk for service providers and system integrators. These blueprints contain a reference hardware and software configuration that has been tested by HPE with support from Red Hat to meet the stringent requirements of CNFs. Examples include HPE’s Telco Core Blueprint C13 Technical Solution:
- HPE Telco Core Blueprint C13 designed to host 5G core CNFs on Red Hat OpenShift in centralized data centers
- HPE Telco Edge Blueprint E8 designed to host RAN Distributed Unit (DU) CNF on single node OpenShift at the far edge of the network (closest to radio units)
- HPE Telco Edge Blueprint E9 designed to host CNFs as well Multi-access Edge Computing (MEC) applications in near edge locations
A network vendor is currently using the laboratory to test a CNF prior to its deployment at a mobile telco service provider in APAC. As some key features become available in the upstream Kubernetes ecosystem, we also expect the laboratory to be used to test and demonstrate sustainability features such as increased power efficiency.
Red Hat invites service providers, NEPs Network Equipment Providers, ISVs and System Integrators to participate in the Red Hat APAC 5G and Edge Technology Lab to test CNFs and develop 5G and edge solutions using Red Hat OpenShift.