I’ve seen a couple of posts (of course they were chock full of AI slop images) on LinkedIn in the last couple of weeks, talking about how challenging it is to implement Kubernetes. In the most recent post I saw, it stated that “it took 5 months for our CEO to implement Kubernetes for our app”, to which I would ask, why the hell is your CEO configuring your clusters. I designed, and implemented the Kubernetes infrastructure on my current project, and I’ve worked on for a while, so of course, I felt the need to share my opinions on the matter.
If you are trying to build bare metal Kubernetes (are you also compiling your own Linux?), it is probably pretty difficult. If you are like the rest of the word, just use your preferred cloud providers Kubernetes distribution (Azure Kubernetes Service or the Elastic Kubernetes Service on AWS) and run with that. Even if you suck at AWS security like I do, you can get this up and running in a couple of hours. I’d even say you could run it on-prem on VMware, but Broadcom threatened me with a lawsuit for saying that without a license. After that, you really don’t have to think about Kubernetes that much other than deploying containers to run as pods.
Yes, this does mean your developers have to learn YAML (or ask an LLM to make it for them), which they should already know, understand how containers works (which I hope they do already), and learn a few organizing things about security and labeling in K8s. But after that Kubernetes handles auto-scaling (especially if you checked the auto-scale box in your cloud provider), does a lot of heavy lifting for networking in your microservices app, and provides a pretty good level of high availability.
Kubernetes has about the same level of complexity as your average cloud deployment, and the infrastructure as code scenarios are far simpler. If you just think of it as VMware, but for containers, and roll with changes, you’ll have a good time and gain a lot of functionality for a bit of work.