Two Aspects of Edge Compute to Focus on for Reducing Edge Complexity

March 8, 2022

As organizations look to capitalize on the benefits of edge computing, many are quickly realizing the complexities associated with building and operating distributed systems – including sourcing distributed compute, resource management/placement/sizing/scaling, distributed traffic routing, managing many locations and non-homogenous providers – leading them to seek out solutions to help solve the edge puzzle. It’s important to understand these complexities, which is why we’ve discussed them in great detail in our previous blog posts. But as with any endeavor, when faced with adversity or confronted by hurdles, keeping an eye on the prize can ultimately help you forge ahead.

Every organization, team and application has varying and distinct requirements when it comes to designing distributed systems. Given that, it’s not uncommon for teams that start down the path of building their own bespoke systems to quickly find themselves overwhelmed by all the complexities that play into design decisions and implementation. As an Edge as a Service (EaaS) provider, we’re often pulled into projects during the early stages of research and discovery, where we’re able to offload the build and management of many, if not most, of the critical components. As such, we know first-hand that accelerating the path to edge for organizations – across a diverse range of use cases – begins with reducing complexity from the outset. Moreover, providing application developers with the opportunity to seamlessly distribute and run the software of their choice is where the real power of the edge comes into play.

To that end, let’s shine a spotlight on a few critical areas where complexity can creep into your edge deployment scenario, and how using an “as a service” strategy can combat that to reduce edge complexity.

Deployment Flexibility (How and Where)

Organizations moving to the edge should expect greater flexibility in both how and where they deploy to edge resources compared to using legacy CDNs or centralized cloud environments for edge. The two areas to consider here are network provisioning and deployment pipelines – in other words, how are your getting your application workload to the edge and where is the edge? Both represent areas where unintended complexity can stymie edge outcomes.

The Where

If edge compute is appropriately provisioned and distributed across a heterogenous network of providers, not only do you get increased flexibility, it also ensures increased resiliency. The Section Composable Edge Cloud, for instance, is built on the foundations of providers such as AWS, Azure, Digital Ocean, Equinix, GCP, Lumen, and RackCorp; we are adding new edge location providers and can deploy points of presence (PoPs) on-demand to help customers define their own edge.

Yet managing such a federated multi-cloud and edge network can increase complexity exponentially. This is where our Adaptive Edge Engine again saves the day by using a sophisticated decision and execution engine to automate distributed infrastructure provisioning, workload orchestration, scaling, monitoring, and traffic routing.

The How

Deployment pipelines also need to be flexible to keep pace with technology changes and application workload demands. Engineers must be able to deploy fast, easily, and in a safe, reliable and repeatable way. DevOps and continuous delivery (CD) help to support a more responsive and flexible organization that can better respond to changing requirements, and ensure quicker time-to-market across the software delivery cycle. That’s why Section leverages your existing Kubernetes managed application structure – no need to rewrite your application to get it to the edge. What’s more, you’re not locked into specific application providers – feel free to use any containerized application from any registry (be it open-source or private).

Management Control

Once your application is at the edge, a modern EaaS provider should enable greater granular control over your tech stack and edge computing requirements. That control allows for better tailoring of application workloads, allowing you to focus on customer or team needs and leverage EaaS for deployment, rather than rearchitecting or retooling to fit the requirements of a cloud or CDN platform. EaaS solutions like Section’s Edge Platform provide the necessary granular, code-level control needed to seamlessly integrate edge solutions into existing stacks and workflows.

Code configuration and management in particular are areas where complexity can impact the developer experience. Beyond the granular, code-level control over edge configuration mentioned above, there are also considerations in terms of location strategy, security and compliance, application development lifecycle, and observability.

At Section, our approach is to combine that fine-grained control when and where you want it, with powerful automation and underlying AI to hide and abstract that complexity where you don’t. For example:

  • The Kubernetes Edge Interface (KEI) provides a K8s consistent interface to deploy and manage workload on the Section Global Network. Working with KEI is analogous to working with K8s to deploy to a single K8s cluster but will actually deploy to Section’s “Global Cluster of Clusters” on the Global Network pursuant to deploy Policy you prescribe via the KEI.
  • Adaptive Edge Engine delivers superior performance/cost efficiencies, including optimal cost to latency – so you don’t end up overpaying for underutilized (“always on”) resources or under serving customers. It also enables direct control of max cost – to liberate the fear of paying more than budget constraints in any given month by keeping costs in check. Performance-wise, Adaptive Edge Engine also allows you to specify the network shape that suits your users – eliminating constraints related to compliance or specific providers.

Abstracting Edge Complexity

Section was founded with a mission to ensure edge simplicity and has thus built a very different type of network to legacy CDNs and cloud providers. That starts with offering an OpEx model for edge compute, allowing us to use flexible strategies and workflows to best meet the needs of each individual customer, maximizing performance and cost savings. Then we simplify network provisioning and deployment, so you can get to the edge fast, without having to change your development cycles, process or tooling. And without having to worry about managing and adapting to a range of different operators. Then we give you the granular control – not to mention support – needed to optimize your application workloads at the edge, and automation that ensures you don’t have to dive deep into the edge to improve overall customer experience.

The Section Edge Platform offers organizations the ability to benefit from the expertise and resources of an EaaS provider who can provide turnkey solutions for customizing and managing these complex systems. By abstracting the complexities of edge computing, organizations can focus on delivering better applications, not operating distributed networks.

Let us help you realize these edge outcomes quickly and easily so you can enjoy all the benefits of a dynamic, customized Edge – at the same cost as cloud – without sacrificing DevOps simplicity, control, or flexibility.

It’s that simple. :)

Similar Articles