There are great tools like Helm which make deploying apps in a kubernetes cluster very easy. A kubernetes cluster groups one or more nodes to run the workloads needed by your apps. Usually these nodes or machines, end up being in the same geographical region, for better connectivity among them. Which means, if one region goes down, all nodes in that region in that cluster could go down too. So how do you ensure disaster recovery to keep your app running in such a setup? Usually most setups use multiple clusters. But although launching an app in one cluster with helm is easy, launching & managing an app across multiple clusters isn’t that straightforward yet. For this, we have created a project called Multi-Cluster App. It’s a light way of pushing the app to multiple clusters. Plus you can use your existing manifests to do so. Some of it’s benefits include:
- Disaster recovery for your app
- Reduced Latency
- Extends Kubernetes, hence it’s a great example for how k8s CRDs and controllers are writtern.
In this session, I’ll discuss this opensource project in depth