As with many things, COVID-19 sped up the need for high bandwidth delivered fast to end users. Working from home, Zoom calls, VPN connections, social media—all of it requires speed, resiliency and robust IT foundations. While Kubernetes is young and still maturing, the one place it is most needed is 5G deployment and scaling.
A resilient and fault-tolerant far edge, fast near edge and resilient core are the three components required to bring 5G to the consumer. Kubernetes meets all of those needs with its agility, rapid deployment and recovery and extendable architecture. It allows operators to reap all the benefits of new, 5G technology and, most importantly, delivers functionality to users at breakneck speed.
Let’s break down each part of the necessary 5G structure:
Far edge: whether it is rugged boxes sitting under a physical tower or a centralized hub point connecting multiple endpoints, the need for fault tolerance, collection and processing of data at speed (and the ability to operate disconnected from the core, if necessary) are all key to a successful implementation of the new cloud-native service-based architecture (SBA). Cloud-native applications are able to recover quickly from any combination of failures at any time, without loss of service or human intervention.
Near edge: the middle layer of the system and maybe the most important of the 3, this is where we strive for speed, low latency, fast aggregation of multiple data streams and returning that data back out for consumption to the Far Edge nodes. Understanding and finding bottlenecks, finding new paths of service, creating analytics or scanning for vulnerabilities, all require accelerated networking and storage, with microsecond level network and storage latencies.
Core: This is where multiple technologies and layers need to be able to mesh and interconnect in a single pane of glass. Consider it the heart of the whole system with a global overview of multiple instances, multiple providers, deeper analytics and where long term storage all work together.
Let’s discuss how companies can manage all these layers:
Far edge: Full-stack Kubernetes solutions for the hybrid cloud with integrated storage and networking simplifies scaling, boosts security by providing all-in-one Kubernetes in a box. Service will continue as long as one instance of each required microservice is available. The degradation of capacity can be addressed by automated scaling, enabling a system to recover quickly and automatically. Providing enough performance for a disconnected node to do local processing until it is able to establish connection further up the chain, while also maintaining local operability through multi-hardware architecture.
Near edge and Core: Organizations will need something akin to a global management plane across edge, near edge and core, allowing the operations team a deep view into infinite scale deployments. It also supplies the mechanisms with which to observe & control such deployments, delegate responsibilities for individual components or whole sections of deployments.
When all of these requirements are combined and compared to what Kubernetes can deliver on its own, it is a near-perfect match. Where Kubernetes leaves the end user out to dry is when it comes to providing anything but a general API for third-party providers to address latency and disaster recovery. It’s important to fill this gap by combining Kubernetes solutions with Cloud Native Computing Foundation (CNCF) Certified Kubernetes distributions and best-in-class private (or public) cloud management offerings. This empowers end users to rapidly take advantage of the exciting new features Kubernetes brings to the table.
More and more providers are delivering their 5G implementations in containerized form for existing telecom companies to deploy and know that they will be able to take advantage and leverage all of the exciting new technological breakthroughs Kubernetes can power.