Possibly linux

You update and then the entire system breaks (because Arch)

Neither of those is going to happen…

You also could do nslookup on Windows

Not necessarily.

A i5-6500 has a TDP of 65W while a i5-13600K has a TDP of 150W.

If you get something modern that has the performance of a i5-6500 it will be a little bit more efficient. The key is that more performance uses more power.

Raspberry Pis are way overhyped and overpriced.

Also this is totally wrong. Once you start it just keeps growing unless there is some other factor.

I personally wouldn’t do this. You want your network to be dedicated hardware

RIP

I don’t think you are going to find anything faster. Ollama is pretty much as fast as it gets

Ansible

I don’t want to spend a bunch of time troubleshooting something. Having a way to move my stuff to a different host when the host crashing is very nice.

What hardware are you looking at?

I would do a three node cluster (maybe even 5 node)

It isn’t to crazy to install

Did you learn something?

I wouldn’t trust his guides personally. He has some hot takes and more importantly he isn’t someone who really knows the Homelab/self hosting landscape.

If you are looking for guides I would find channels that have done series on whatever you are interested in there is plenty of quality material.

To start off here is what I would do.

First, get a wireless router that is capable of running OpenWRT and then get a switch to accompany it.

Next go to eBay and buy 3 used workstations. They don’t need to be fancy and you can always upgrade them later. You need 3 for later.

Next find some storage. You can find decent Sata SSDs for pretty cheap. If you are looking to store something bigger like a movie collection also pickup some larger drives. With the extra drives make sure you buy a sata or SAS pcie card. This is because you need a dedicated controller to passthough to a VM.

Once you have all that you can start installing Proxmox. You probably want a raid 1 configuration so that you can replace a disk without downtime. The reason I say three devices is because you need 3 machines to get consensus in the cluster. When consensus is lost affected devices go into what is called fencing which is where it freezes all VMs and operations to prevent split brain from happening.

Technically this is probably a bit overkill but I like having a solid base for experimentation and flexibility. Doing it right from the get go will mean that you have more power down the road.

For actually hosting stuff I would use docker compose inside a VM.