By W. Brooke Frischemeier, Senior Director of Product Management, 5G and Edge at Robin.io

The technology revolutions of the past two decades have taught us that time to market, solution flexibility and user experience are each critical components for success. We see this repeated with the advent of 5G, where the needs of 5G core, RAN and MEC services are more diverse than ever and require high performance with both rapid and immense scale while reducing integration timelines, resources and operations silos. 

A key decision operators now face when it comes to accelerating time to market, in a vibrant multi-vendor 5G services ecosystem, is choosing orchestration and cloud-native platforms that enable a rich set of vendors, applications, deployment environments and lifecycle demands. While vendor diversity is important for long term continued growth and costs, it can be a double-edged sword. The broader an ecosystem, the more flexibility and business opportunities await. However, this can come at the cost of deployment timelines and operational complexity.

In many cases, vendors have their own orchestration, lifecycle automation and even Kubernetes platforms. When we put them all together, it becomes a challenge to manage multiple tools and the necessary interoperability in a way that promotes common workflows across multiple hardware and service domains. Because of this, there is a great need for a unifying cloud platform, orchestration and lifecycle automation layer supporting network functions, services, stateful applications, appliances and the underlying hardware infrastructure. Doing this efficiently means delivering all of them over a single pane of glass used to plan, operate, manage and visualize 5G services, while supporting zero-touch provisioning and lifecycle management. Otherwise, future 5G demand will outstrip the ability to adapt and scale efficiently, limiting growth potential.

Focusing on the ‘how', to meet the ‘when'

Meeting detailed 5G service requirements across so many use cases and industries is both a business opportunity and a serious challenge. Many of the new services promised by 5G are still incubating.  As we move closer to implementation, much of the industry spends a lot of time discussing the "what" checkboxes that platforms and tools tick off instead of "how" they reduce the "when" of a deployment. Platforms and automation tools need to take the difficult tasks-integration, deployment, lifecycle management-and reduce those timelines. Otherwise, they will quickly become a bottleneck. 

This transformation is becoming increasingly difficult for incumbent platforms with bolt-on features and complex licensing schemes, as they were built long before the cloud-native and 5G revolutions. Fundamentally, the design of 5G  solutions needs to change and become more dynamic than previous generations, embracing an expanded ecosystem with reduced timelines as its core principles. The question is, how do we get there and how do we avoid potential implementation dead-ends? It's by asking "how" that sound technology choices can reduce the "when" of deployments.

Transformation at such a scale has never been a forklift upgrade. Furthermore, one needs to be able to retire incumbent assets when the numbers make sense. Automation tools and cloud-native platforms must be capable of bringing embedded hardware, software and network functions forward independently, otherwise operators are permanently tied to someone else's modernization roadmap. To accomplish this, tools and platforms must support more than just a single paradigm (for example, VMs and containers) as well as existing automation executors and multiple hardware platforms.

This inclusiveness is even more important in resource-constrained edge data centers that also provide the additional challenge of being remote. Slapping a container or bare metal add-on, complete with new licenses, to an incumbent platform does not guarantee the removal of resource and operations silos because it was never designed to do this. In fact it usually means added complexity that reduces implementation timelines putting capacity planning and expansion at risk.

Another important way to streamline operations is to adopt a platform that seamlessly integrates with bare metal operations. Why shouldn't operators be able to unify bare metal provisioning, cluster deployment, networking and services into a single, context-aware workflow that doesn't require manual baby-sitting for every single step? It's supposed to be automated! How easy is it to add, clone and migrate applications from the core to remote and often different hardware environments at the edge and far edge cloud.

Next, we can improve upon scalability timelines by digging a bit deeper into existing tools such as Helm, which provide a useful paradigm for application onboarding. However, as we scale for 5G, it can easily become a scripting rats nest, slowing down automation timelines, vendor onboarding and solution adaptability. Ask what your platform does to not just reuse existing paradigms such as Helm, but enhance flexibility, network function integration, ease of use and automated workload placement. Furthermore, ask how platforms transcend legacy lifecycle tools and provide workflow tooling that behaves more like application and service workflows? How does it change what was once complex into something that is intuitive and streamlined? How does it reduce the time it takes to onboard new vendors?

The pay-off

The full impact of 5G applications is still yet to be realized. First-adopters that embrace the aforementioned principles can leverage a rich vendor ecosystem promoting the rapid deployment of flexible solutions, entering new markets faster and staying ahead of nimble competitors because they themselves have become the nimble competitor! By combining bare metal to services workflows over both virtual machines and containers on the same Kubernetes platform, operators can take advantage of a unified operations model and fully shared resource pools, unlocking the limitless potential of a rich 5G ecosystem, creating new business revenue, saving costs and reducing time to market, getting there first!

##

To hear more about cloud native topics, join the Cloud Native Computing Foundation and cloud native community at KubeCon+CloudNativeCon North America 2021 - October 11-15, 2021       

ABOUT THE AUTHOR

W. Brooke Frischemeier, Senior Director of Product Management, 5G and Edge at Robin.io, a KubeCon NA 2021 sponsor. Visit them at booth #S84.

Brooke is Sr Director of Product Management for 5G and Edge at Robin.io. Brooke brings close to three decades of extensive experience in the telecom & networking domain and has held leadership positions in organizations like NetNumber, World Wide Technology, Cisco, FORE Systems, and Bell Laboratories.  His experience spans Product Management, Business Development, Partner Management, System Engineering, and Professional Services. His expertise & interest include container orchestration, virtualization, several generations of mobile networks, switching routing, and telephony across Enterprises and Providers.