Kubernetes adds a lot of complexity. In return, it allows various teams in your company to work mostly independently, so that your software stack can mirror your org chart better. It trades latency for scalability (adds network calls to things that could have been local function calls). If your “home lab” isn’t serving millions of users, you don’t need Kubernetes to run it.
That said, you might be using your home lab partly as practice for a job at a large company where the tradeoffs of Kubernetes make sense (or at least someone thought they made sense and started using it, which is more common). That means using it at home can provide valuable self training, since you can screw around and not take down the production cluster for anyone other than yourself.
Kubernetes adds a lot of complexity. In return, it allows various teams in your company to work mostly independently, so that your software stack can mirror your org chart better. It trades latency for scalability (adds network calls to things that could have been local function calls). If your “home lab” isn’t serving millions of users, you don’t need Kubernetes to run it.
That said, you might be using your home lab partly as practice for a job at a large company where the tradeoffs of Kubernetes make sense (or at least someone thought they made sense and started using it, which is more common). That means using it at home can provide valuable self training, since you can screw around and not take down the production cluster for anyone other than yourself.