If you run only system resources, or task manager, or whatever windows is calling their resource manager these days to monitor CPU, right next to a headless debian server running only htop you will straight up see the answer to your question.

That, is overhead.

:) But you have a system in place should you need it. And I say All Data can be useful if filtered correctly.

Did it help?

Oh awesome! I love finding additional uses for my pi’s.

Wow. Thorough. I like it! 👍

I had to upvote this because I like the added “progression path analysis” given. Everything checks out so far from my personal experience.

However I have not yet delved into kubernetes yet.

Could the poster of this reply elaborate (briefly is fine) what some advantages are with Kubrn8s? You mention redundancy. From my completely inadequate understanding of kuber, you can cluster together the resources of different individual systems? Like how truenas can use all the storage of different sized drives to form one pool that can be managed as 1 resource? This of course would just be an example of what it does in concept?

So theoretically, one can sort of network a cluster of old PC’s to make a really decent, redundant “server” that shares the workload?