Kubernetes use cases

As we have already touched upon in this chapter, Kubernetes can run pretty much anywhere, from just your local machine (which we will cover in our next chapter), from your on-premise hardware of virtual machine infrastructure to potential spanning hundreds of public cloud instances in AWS, Microsoft Azure, or Google Cloud. In fact, you could even span multiple environments with your Kubernetes cluster.

This means that you get a consistent experience no matter where you are running your application, but also get to take advantage of your underlying platform's features, such as load balancing, persistent storage, and auto scaling, without have to really design your application to be aware it is running on, say, AWS or Microsoft Azure.

One of the common threads you will notice when reading through success stories is that people are talking about not being locked into one particular vendor. As Kubernetes is open source, they are not locked into any licensing costs. If they have a problem or want to add functionality, they are able to pe straight into the source code and make changes; they can also contribute any changes they make back to the project via a pull request.

Also, as already discussed, using Kubernetes allows them to not get locked into any one particular platform vendor or architecture. This is because it is reasonable to assume Kubernetes will perform in exactly the same way when installed on other platforms. Because of this, all of a sudden you are able to take your application and move it between providers with relative ease.

Another common use case is operations teams using Kubernetes as an Infrastructure as a Service (IaaS) platform. This allows them to offer their developers resources they can consume via APIs, the web, and CLIs, meaning that they can easily hook into their own workflows. It also provides a consistent environment for local development, all the way from staging or user acceptance testing (UAT) to eventually running their applications in production.

This is part of the reason why using Kubernetes to execute your serverless workloads is a good idea. You are not locked in by any one provider, such as AWS or Microsoft Azure. In fact, you should think of Kubernetes as a cloud platform like the ones we looked at in Chapter 1, The Serverless Landscape;  it has a web-based console, an API, and a command-line client.