Now Available: AI/ML Industrial Edge Solution Blueprint v1.0

Update: This blueprint has been updated to version 2.0. Find out about the new version here.

Many industries, including industrial/manufacturing, are bringing together the powerful combination of edge computing and AI/ML to transform their operations and fuel innovation faster by bringing processing power closer to data. In our previous blog post "Boosting manufacturing efficiency and product quality with AI/ML, edge computing and Kubernetes", we explained how and why someone would use OpenShift at the edge on a factory floor, we have now released version 1.0 as a complete GitOps repository which everyone can use, study and even contribute to. In the repository you will find:

• Comprehensive documentation
• Configuration for the central cluster
• Configuration for the clusters deployments in the factories
• Example workloads GitOps repo
• And the Jupyter notebook for Machine Learning workload

We created the blueprint with a few key goals in mind:

• First, we wanted to show how Red Hat’s comprehensive portfolio could be used to address an edge use case (in this case, the industrial edge)
• Second, we are using solution blueprints internally in our CI to verify that integrations which are working one day, keep on working over time. This is an area where many of our customers have asked us for help and now, via the work done in bringing this blueprint together, we will be able to identify any integration issues earlier in the release cycle of our products.
• Third, we wanted to go beyond the traditional reference architecture which is usually a long list of "printed" instructions. The GitOps model enables us to deliver the blueprint code in a way that is readily available to our customers and partners. What this also enables you to do, is to easily use this blueprint for a POC, modified to fit a particular need and hopefully transformed into a real deployment.
• Finally, our blueprints are open source, so anyone can suggest improvements, contribute to them, or fork them to do something else.

For this particular solution blueprint, we demonstrate how OpenShift, ACM, AMQ Streams, OpenDataHub, and other Red Hat products come together to address an edge computing use case commonly found in manufacturing: Machine inference-based anomaly detection on metric time-series sensor data at the edge, with a central data lake and ML model retraining. You can watch a full demonstration of this which we recorded for an OpenShift Commons briefing.

We really hope to see you on our Git soon!

To learn more, visit www.openshift.com/edge