Deploying Applications

Remoteness

Deploying resources to a Kubernetes cluster running in AWS is no different from deployments anywhere else, including Minikube. That’s one of the big advantages of Kubernetes, or of any other container scheduler. We have a layer of abstraction between hosting providers and our applications. As a result, we can deploy (almost) any YAML definition to any Kubernetes cluster, no matter where it is.

It gives up a very high level of freedom and allows us to avoid vendor locking. Sure, we are cannot effortlessly switch from one scheduler to another, meaning that we are “locked” into the scheduler we chose. Still, it’s better to depend on an open source project than on a commercial hosting vendor like AWS, GCE, or Azure.

ℹ️ We need to spend time setting up a Kubernetes cluster, and the steps will differ from one hosting provider to another. However, once a cluster is up-and-running, we can create any Kubernetes resource (almost) entirely ignoring what’s underneath it. The result is the same no matter whether our cluster is AWS, GCE, Azure, on-prem, or anywhere else.

Deploying Resources

Let’s get back to the task at hand and create go-demo-2 resources.