The open source community signaled its intention to dominate edge computing during a KubeCon + CloudNative + Open Source Summit China conference this week in Shanghai.
Arpit Joshipura, general manager for networking, automation, edge, and IoT at the Linux Foundation, said that, for all intents and purposes, open source platforms such as Kubernetes have already won the battle.
For example, Huawei Technologies demonstrated at the conference how it is leveraging the open source Kubernetes to extend the reach of its cloud services to the edge of the network.
The HUAWEI CLOUD is built on an Intelligent EdgeFabric (IEF) that enables Huawei to deliver services to multiple platforms, which includes a Huawei KubeEdge offering based on Kubernetes that is optimized for edge computing applications that have access to limited memory. That instance of KubeEdge not only connects to HUAWEI CLOUD; it can also run what are known as Kubernetes pods in an offline environment.
Joshipura told conference attendees that approach to edge computing will be copied by most service providers because the cloud is now essentially now moving to the edge. To facilitate that transition, the Linux Foundation has formed LF Edge, an arm of the Linux Foundation that is building a set of frameworks and validated reference architectures for build edge solutions, for everything from Internet of Things (IoT) environments to 5G wireless networking services based on a Akraino Edge Stack platform that has Kubernetes at its core.
The impact of LF Edge
There are already more than 60 vendors that are members of LF Edge, including AT&T, Dell EMC, General Electric, IBM, Intel, and Samsung. LF Edge has already created a set of reference architectures for implementing Akraino Edge Stack and will soon create blueprints for implementing that platform with specific vertical industry segments.
Edge computing in all its forms is the next trillion-dollar IT industry. Everyone from open source communities like LF Edge, to providers of proprietary platforms, such as Cisco and VMware, are squarely focused on this emerging opportunity.
LF Edge is essentially making the case for employing the same open source technologies that are just now starting to dominate the next generation of cloud computing at the network edge. The expectation is a common framework running at both the edge and the cloud will facilitate the development of hybrid cloud computing environments that can truly span multiple clouds.
Of course, everyone’s definition of the edge is different. The Akraino platform assumes there is an instance of Kubernetes running on some type of hyperconverged infrastructure (HCI) platform that functions as a gateway. In turn, that gateway is connected to all kinds of systems that may not have enough local compute resources to run Kubernetes.
Regardless of the approach, the edge computing platforms are already rapidly being commodity compute platforms. The implications of that for managed service providers (MSPs) is profound because it means they can craft a service using any number of “white box” appliances. The truth is most end customers are not going to be especially interested in what infrastructure is used to deliver an edge computing service.
Put it all together and it is clear edge computing is not likely to be as dominated by big brands the way traditional data centers are today. Instead, it will be the quality of the edge computing experience it enables that will determine which service provider ultimately rules the edge.
Photo: Matthew Brodeur / Unsplash