After covering the business drivers and the deployment challenges for edge computing in the first two blogs in my Getting to the Edge blog series, I will now discuss how to pull it all together in this final blog in the series, as well as look at specific use cases.
When it comes to delivering edge computing solutions, we will see a new class of systems integrators emerge that can deliver a complete and fully integrated solution with zero vendor lock-in. From Radisys’ point of view, leveraging open source technologies is key for MEC, and one of the most important ways that service providers can accelerate new services. We advocate that our customers take a building block approach that leverages third-party components from vendors such as Radisys or others, MEC applications, MEC platforms and orchestration.
Figure 1: The Building Block Approach
The goal of this approach is reduced time-to-market, through validation of deployment and location choices, while enabling our customers to bring MEC solutions towards production quickly.
Our Edge Computing Platform Software is one such integrated solution which provides an access-agnostic, ETSI standards-compliant mobile edge platform with a building block framework and open APIs to enable disruptive edge applications. It is based on Intel® Xeon® processor-based platforms and leverages Intel’s Network Edge Virtualization (NEV) SDK targeted at MEC applications and services. Our platform can be used with multiple access technologies, enabling service providers to leverage a common edge computing platform for both their wireless and wireline solutions, and to break vendor lock-in from proprietary edge computing solutions.
A Look at Use Cases / Mobile Edge Applications
There are a number of applications that benefit from an Edge strategy and platform built with open building block components.
Figure 2. Demonstrating Leading Edge of Open Mobile Innovation
For example, Real-Time Media Breakout in venues with a large physical location such as a U.S. football stadium, allows service providers to offer the option to see what is happening in other areas of the stadium with the right levels of policy control for the media based on RAN conditions. With this use case, service providers can increase revenues by offering personalized consumer viewing experiences, and reduce costs and improve the user experience by lowering backhaul costs due to reduced amount of traffic sent to the core network and reducing the latency for each user viewing request.
For Video Analytics, you take the video stream and apply analytics at the edge. This goes back to the Laws of Economics and Physics covered in the first blog in this series where there is so much data coming in, but the data may be of low value, except for event-based triggers that are implemented such as triggers of audio and video feeds. Service providers can use video analytics for use cases such as facial recognition, camera tampering detection, and people or vehicle detection and counting.
Finally, Firmware OTA based on RAN capabilities and conditions can enable over-the-air firmware upgrades – such as for parked cars – in a reliable way. You can apply policy enforcement to prevent upgrade failures. Benefits include augments scalability by adding new features to products after they are released; continuous improvement of solutions; companies can test new features by sending updates to one or multiple devices; and developers can deploy frequently, knowing that products stay functional as updates are released. Service providers can also save on firmware transport costs by deploying edge solutions.
How Does the Solution Work? Integration and Customization
The edge solution is comprised of a complete set of functional building blocks, which are pulled together and integrated by Radisys. We then provide this solution as a reference design to our customers. Not every customer needs the same set of building blocks, so we customize based on each customers’ requirements and use cases. Below is an example of the major building blocks that are used in an edge solution.
Figure 3: Integrated Solution Functional Blocks
We use building blocks from open ecosystems such as ONF’s M-CORD (Mobile Central Office Re-Architected as a Datacenter) platform, we use ONAP for orchestration, and we use xRAN/ONAN controller for the eNodeB control. We are active in these groups to drive network transformation.
One of the new ecosystems that’s been created to focus on Edge deployments is the Linux Foundation Akraino project (https://www.akraino.org/). This new ecosystem is focused on developing an open software stack optimized for Edge computing applications. Radisys is a founding member of Akraino and plans to participate in two production Edge deployable blueprints (with ARM and another with RedHat, Intel). The first version of the Akraino stack along with the supported blueprints are expected to be available 1H-2019.
At Radisys, we are focused on taking the pain out of open integration for our service provider customers and on delivering complete integrated Edge Computing Solutions that meet each customer’s needs. Please contact us at firstname.lastname@example.org to learn more or to schedule a meeting with us at Mobile World Congress and meet with our edge experts.
About the Author