Build a Kubernetes "Intelligent Edge" Cluster on Raspberry Pi

[glovebox]

Follow me on Twitter @dglover

Introduction

Building a Kubernetes Intelligent Edge cluster on Raspberry Pi is a great learning experience, a stepping stone to building robust Intelligent Edge solutions, and an awesome way to impress your friends. Skills you develop on the edge can be used in the cloud with Azure Kubernetes Service.

The Kubernetes cluster is built with Raspberry Pi 4 nodes and is very capable. It has been tested with Python and C# Azure Functions, Azure Custom Vision Machine Learning models, and the NGINX Web Server.

This project forms the basis for a four-part Intelligence on the Edge series. The followup topics will include:

• Build, debug, and deploy Python and C# Azure Functions to a Raspberry Pi Kubernetes Cluster, and learn how to access hardware from a Kubernetes managed container.

• Developing, deploying and managing Intelligence on the Edge with Azure IoT Edge on Kubernetes.

• Getting started with the dapr.io, an event-driven, portable runtime for building microservices on cloud and edge.

System Setup

Get Started

The Kubernetes cluster set up is fully scripted and well documented.

Head to Building a Kubernetes Intelligent Edge Cluster on Raspberry Pi on GitHub to learn more.

[James van den Berg]

Thank you for this Cool Blogpost !

I have shared this with Education in the Netherlands  for the Students.

[glovebox]

Hey James, thanks for the feedback, it was a lot of fun to build and it almost broke me, but the project and the installers have now stabilized, phew:) There are a lot of opportunities to extend the project. Let me know how the students get on and please encourage them to post issues on the GitHub page. Cheers Dave