Building a Kubernetes "Intelligent Edge" Cluster on Raspberry Pi

Building a Raspberry Pi Kubernetes Cluster

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.

Learning Kubernetes

You can download a free copy of the Kubernetes: Up and Running, Second Edition book.

It is an excellent introduction to Kubernetes and it will accelerate your understanding of Kubernetes.

Published: 8/22/2019

Improve the agility, reliability, and efficiency of your distributed systems by using Kubernetes. Get the practical Kubernetes deployment skills you need in this O'Reilly e-book. You will learn how to:

• Develop and deploy real-world applications.
• Create and run a simple cluster.
• Integrate storage into containerized microservices.
• Use Kubernetes concepts and specialized objects like DaemonSet jobs, ConfigMaps, and secrets.

Learn how to use tools and APIs to automate scalable distributed systems for online services, machine learning applications, or even a cluster of Raspberry Pi computers.

Azure Kubernetes Services

Kubernetes Resources

Installed Demos

Before run the demos you need to insure an instance of the corresponding demo pod is running. By default, there are zero pod instances active when the cluster is set up. The easiest way to start a pod is from the Kubernetes Dashboard Deployments and Replica Set tabs

You need the application service address and port. Again. most easily obtained from the

• Kubernetes Services tab
• or from the command line:
  
  kubectl get services

  NAME               TYPE           CLUSTER-IP      EXTERNAL-IP       PORT(S)          AGE
  image-classifier   LoadBalancer   10.96.136.29    192.168.100.200   80:32519/TCP     4h25m
  jupyter            LoadBalancer   10.96.224.173   192.168.100.203   8888:30471/TCP   4h25m
  kuard              LoadBalancer   10.96.229.114   192.168.100.201   8080:32515/TCP   4h25m
  kubernetes         ClusterIP      10.96.0.1       <none>            443/TCP          4h26m
  led-controller     LoadBalancer   10.96.148.9     192.168.100.205   80:30752/TCP     4h25m
  nginx-service      LoadBalancer   10.96.9.213     192.168.100.202   80:32711/TCP     4h25m
  openweathermap     LoadBalancer   10.96.109.92    192.168.100.204   80:31732/TCP     4h25m
  

Preinstalled Demos

1. NGINX Web Server: http://{nginx-service EXTERNAL-IP}
2. Jupyter Notebook (Python3, Pandas, Numpy, Scipy, Tensorflow 2): http://{jupyter EXTERNAL-IP:8888/}
3. Led Controller: http://{led-controller EXTERNAL-IP}/api/ToggleLed?led=0
4. Image Classification: curl -X POST http://{image-classifier EXTERNAL-IP}/image -H "Content-Type: application/octet-stream" --data-binary @banana.jpg

Running Demos

Image Classification

LED Controller