How 5G Networks will Influence the Location of Edge Compute


By Bill Wagner | February 26, 2019 | Bright Cluster Manager, edge computing, 5G



While the world eagerly awaits the rollout of 5G networks and the exciting new apps for both consumers and businesses that will be made possible via 5G services, enterprises, telcos, hosting providers and others are scrambling to architect and build the computing infrastructure necessary to support the new requirements of these apps. Two key challenges in defining the computing infrastructure for 5G apps are that they will be data/compute intensive and have little tolerance for network latency. The former means that significant computing capacity is needed to deliver the intended experience of the app and the latter means that the compute must be located in relative proximity to the user of the app. The result: A significant build-out of localized compute (predominately Linux servers and clusters) will be required to service the new generation of 5G apps speeding towards us. But where (specifically) will that compute infrastructure reside? 

Countless articles have been written about the fact that the network latency associated with centralized computing infrastructures in enterprises and public clouds is too great to deliver the intended experiences of 5G apps, which in turn has driven interest in edge computing as the solution.  In practice, there may be a subset of 5G applications that can be hosted in a centralized computing infrastructure with no problems.  But looking at the primary use cases that will need compute in closer proximity, we’re faced with the reality that the “edge” is a bit of a moving target. For some applications, a localized public cloud region located in a residential area may provide the right balance of compute and latency coupled with the convenience of a public cloud model. For applications with more aggressive latency requirements, it may be necessary to put the computing infrastructure on-premise within a business. Regardless of where this new edge compute resides, organizations building the compute infrastructure for these new 5G apps face the same challenge: how do we quickly and cost-effectively build the necessary computing infrastructure to host the applications I need to run today as well as tomorrow, with minimal staff, and manage that edge infrastructure remotely? 

Bright Cluster Manager software automates the process of building and managing Linux clusters, eliminating complexity and administrative burdens so that organizations can focus their time and energy on higher-value activities.  Bright software allows organizations to build and manage computing infrastructure that spans their data center, edge locations, and public clouds, creating a single, centrally managed cluster that can host bare metal, VM or containerized applications side by side on the same cluster.

So, while 5G will drive the creation of new computing infrastructure that will materialize in different physical locations to suit different use cases, the challenges that organizations will face in building that infrastructure will be the same.  If you’re interested in learning more about how Bright can make it easier, we’d love to speak with you.