Wrapping Up

Cleanup

Let’s clean up the resources we created during the tutorial.

First of all, let’s destroy the Juju controller we boostrapped for the tutorial.

juju destroy-controller --destroy-all-models --destroy-storage --force spark-tutorial

The spark-streaming, history-server and cos namespaces are automatically deleted when the corresponding models are destroyed. Let’s also delete the spark K8s namespace. This will automatically clean up K8s resources within the namespace.

kubectl delete namespace cos
kubectl delete namespace history-server
kubectl delete namespace spark-streaming
kubectl delete namespace spark

Finally, the S3 bucket that was created can be removed using the AWS CLI as follows:

aws s3 rb  s3://spark-tutorial --force

Going Further

Parts of this tutorial were originally covered in a talk at the Ubuntu Summit 2023, the recording of which is available here on YouTube.

This tutorial covers running Charmed Spark locally using MicroK8s. Running Charmed Spark in MicroK8s locally is limited by the amount of resources available locally. For a more robust deployment, it’s also possible to run Charmed Spark solution on AWS EKS. Please refer to this how-to guide for guidance on deploying and configuring an AWS EKS cluster to run Charmed Spark. Additionally, here is a video demonstration of running Charmed Spark on top of AWS EKS at the 2023 Operator Day demo.