Edge Computing and Akraino

  • Overview
  • Resources
  • Transcript

Get an overview of how edge computing is driving the 4th industrial revolution and how Open Source initiatives like Akraino* contribute to the future of networking.

Hi. I'm Sujata. In this video, I give an overview of industry initiatives to develop a common set of software stacks for the edge. I also cover how open source initiatives, like Akraino* contribute to this goal. Edge is all about squeezing the best performance for your app from the network by placing the resources close to where they're needed.

Data is growing. And the demand for low latency, high-bandwidth network functions is also increasing. Applications need to be able to keep up with user demands in real-time. Fixed function devices, like programmable logic controllers, industrial gateways, or human-machine interfaces are becoming software-defined workloads and are driving the need to virtualize, containerize, and look for the new infrastructure management frameworks. The solution for this is LF Edge.

It is an umbrella organization that aims to establish an open interoperable framework for edge computing independent of hardware, silicon cloud, or operating systems by bringing together industry leaders. LF Edge has a project called Akraino. It is an open-source collection of software stacks that support high availability cloud services optimized for edge computing systems. It is a set of blueprints that are integration projects that pull in components from various open-source initiatives to build optimized, deployable, ready-made stacks for a particular use case.

Here are some of the blueprints to pay attention to. First is the Akraino time-critical edge computer blueprint. It enables workloads, such as industrial controllers, Edge Inference, Edge MI, Edge X Foundry to run on the same compute device. This is done with ethernet time-sensitive networking. It leverages Acorn, Zephyr, and Clear Linux.

This blueprint is suited to take full advantage of Intel's new industrial-focused SLCs. They offer and does time-critical compute technology to enable deterministic compute and ethernet TSN for networking. Next, is the Akraino integrated cloud networking blueprint. It addresses the need of Edge Cloud, telco, and enterprise segments in transforming to the cloud-native space. The edge and cloud differ because of constraints in cost, power, and space.

That edge has to support massive geo-distributed compute locations, which exposes it to security concerns and requires the orchestration deployment model to scale for various use cases. Akraino, ICN Blueprint addresses these challenges for various use cases, such as NFV, distributed analytics as a service, EdgeXFoundry, and VR 360 streaming. It uses Kubernetes as the de facto resource orchestration to deploy various container runtimes like Dockers and Kata containers and VMs, like Virlet and KubeVirt.

It also uses ONAP* for Kubernetes as a service orchestration to manage workloads in various edge locations and leverages various open-source projects, such as Openness and OpenWrt*. The Akraino ICN Blueprint has contributions from open source community members, like Verizon*, Dell*, VMWare*, and Aarna Networks. We are looking for developers to contribute to these blueprints to make a grid app to utilize the edge infrastructure. To contribute and learn more, check out the links provided. Thanks for watching.