Production Kubernetes

Chapter 14. Application Considerations

Kubernetes is rather flexible when it comes to the type of applications it can run and manage. Barring operating system and processor type limitations, Kubernetes can essentially run anything. Large monoliths, distributed microservices, batch workloads, you name it. The only requirement that Kubernetes imposes on workloads is that they are distributed as container images. With that said, there are certain steps you can take to make your applications better Kubernetes citizens.

In this chapter, we will pivot our discussions to focus on the application instead of the platform. If you are part of a platform team, don’t skip this chapter. While you might think it only applies to developers, it also applies to you. As a platform team member, you will most likely get to build applications to provide custom services on your platform. Even if you don’t, the discussions in this chapter will help you better align with development teams consuming the platform, and even educate those teams that might be unfamiliar with container-based platforms.

This chapter covers various considerations you should make when running applications on Kubernetes. Mainly:

• Deploying applications onto the platform, and mechanisms to manage deployment manifests, such as templating and packaging.

• Approaches to configure applications, such as using Kubernetes APIs (ConfigMaps/Secrets), and integrating with external systems for config and secret management.

• Kubernetes features ...