Helping CSPs transition to a cloud-native 5G core
Building a cloud-native 5G core network is not easy. It requires careful planning, implementation, and expertise in cloud-native architectures. The network needs to be designed and deployed in a way that ensures high availability, resiliency, low latency, efficient resource utilization, and flawless component interoperability. CSPs may feel overwhelmed when considering the transition from legacy architectures to an open, best-of-breed cloud-native architecture. This can lead to delays in design, deployment, and life cycle management processes that stall projects and reduce a CSP’s ability to effectively deploy and manage their disaggregated cloud-native network.
Automation plays a critical role in managing deployment and life cycle management processes. Many projects stall or fail due to poorly defined automation strategies that make it difficult to ensure compatibility between hardware and software configurations across a large, distributed network. This is especially true when trying to deploy and manage a cloud platform running on bare metal.
Dell Telecom Infrastructure Blocks for Red Hat are foundational building blocks for creating a Telco cloud that is based on Red Hat OpenShift. They aim to reduce the time, cost, and risk of designing, deploying, and maintaining 5G networks using open software and industry standard infrastructure. The current release of Telecom Infrastructure Blocks for Red Hat supports the creation of management and workload clusters for 5G core network functions running Red Hat OpenShift on bare metal servers.
There are a number of challenges to build and maintain Kubernetes clusters on bare metal to run 5G network functions:
- Ensuring interoperability and fault tolerance in a disaggregated network is not an easy task. Deploying and managing Kubernetes clusters on bare metal requires extensive design, planning, and interoperability testing to ensure a reliable, fault tolerant, and performant system.
- Automating the deployment and life cycle management of hardware resources and cloud software in a bare metal environment can be complex. It involves deploying and updating a fleet of bare metal servers at scale.
- There is a lack of pre-built software integrations specifically designed for deploying Kubernetes clusters on bare metal servers and bringing those cluster configurations to a state where they are ready to run workloads. This means that configuring and deploying Kubernetes on bare metal frequently requires more manual effort to build and maintain the automation needed to manage deployments and upgrades at scale. This manual effort can be time-consuming and add complexity that introduces risk to the process.
- This lack of consistent, easy-to-manage automation to deploy and update the cloud stack to meet workload requirements also make it harder to implement a unified cloud platform across all workloads. This leads to infrastructure silos that limit the ability to pool resources to improve infrastructure utilization rates, which in turn reduces network TCO efficiency.
These challenges are amplified when running 5G network functions, which require low latency and high reliability to meet carrier-grade service level agreement (SLAs). This collaboration between Dell and Red Hat aims to offer a comprehensive solution for CSPs that addresses the challenges associated with building and maintaining carrier-grade cloud infrastructure for 5G core network functions.