Use containers. Start with one device. Check your utilization after you're sure you've hit min and max for each of your services, then figure out if your single device can handle all your services gunning at once. If not, take your biggest service and migrate it to its own device.

Eventually, you might find yourself googling "Kubernetes vs Docker Swarm." When you do that, take a deep breath and decide if upgrading one device is easier than trying to horizontally scale many.

Edit: Words bad. Verbs hard.

Pretty much the tradeoff that you said. Harder to maintain an all in one box since things conflict with each other. That said, it's also harder to maintain 10 devices instead of 2. Usually, you want to segregate your services based on maintenance schedule. Something that you reboot once a year like your router probably shouldn't be on the same device as something that you might reboot every day, like home assistant, if you value your sanity.

Also, virtualization is pretty much dead-end now and will just make your life harder.

In terms of the easiest software available for self hosting, I would use a dedicated router and a dedicated nas, as those are fairly standalone and can be purchased as appliances. Then I would use a single machine with Debian or NixOS, and use it as a Kubernetes or Docker host. (Kubernetes is super easy with k3s and easier to maintain than Docker, but there's a higher barrier to entry as you'd have to write your services with Pod files instead of docker-compose files)

I wouldn't recommend something that tries to do everything, like Unraid, TrueNAS, or Proxmox, as they honestly obfuscate things and make things harder to maintain. Though they can be nice for DIY NASes.

If you're interested in high availability and clustering for a DIY NAS, you could even look into ceph/rook, which is what I'm using for my NAS, but it's like 20x the effort of just having a standard NFS appliance.

I'm no expert; only been dipping my toes in the selfhosted water for a few years.

But my thought process would be all the main stuff on your main server and the redundant instances on a little backup

"Easier" and "simpler" are in the eye of the beholder.

A different way to approach it is to limit the failure domains. If this breaks how sad are you?

I would separate storage from the rest. Networking stuff together may be fine. Home assistant depends on how dependent on it your household is.

This is what I do:

• Stuff that's critical runs on VPSes running Debian stable. Things like my websites, email, authoritative DNS, etc. The VPS providers I use have nicer hardware than me (modern AMD EPYC servers, enterprise NVMe drives in RAID10 with warm spares, 40Gbps networking, etc)
• Other stuff is on a home server running Unraid. It has a Core i5-13500 with a W680 motherboard, 2 x 2TB NVMe drives in ZFS mirror, 2 x 20TB Seagate Exos drives in ZFS mirror for data storage, and 1 x 14TB WD Purple Pro for security camera recordings.
• I have a Raspberry Pi with a few things on it, like a second copy of my recursive DNS server, AdGuard Home (so the internet doesn't break if I need to shut down "the main server").

I was thinking of running several servers at home, but right now I'm just running one main one. I don't have much space and it's running fine for me for now. Power is expensive here. I've got solar power, but I get 1:1 credits for excess solar power, so I'd rather save it for other things.

Services that can utilize the full power of a single machine are quite rare. I have about 15 docker containers in total taking up about 800mb of ram on one of my servers. In reality having multiple can be more complex and harder to maintain, not to mention power efficiency and cost.

Having as much on one machine as possible has efficiency and maintenance benefits since you have less machines to configure. The drawback is that multiple services can add up peak demands and run the machine oom which you can either solve by leaving extra headroom or make them redundant imo.

Someone with more experience than me might have other ideas to add.

I'd say mostly energy savings and CPU usage efficiency

It could be a good idea to move more critical things to a different machine. It's often said that you shouldn't run your router and/or firewall on your main server, but I think there are also security reasons for that.

Or to move those to a low power consumption machine with cheaper hardwRe that are either more resource friendly, or very heavy but it's fine if they can only finish their task over a longer time.

Also, think about how could things go wrong. Have a second DNS and DHCP server (it's difficult to run a secondary DHCP besides the primary, maybe you don't need that), and some way you can reach the internet if the router or the firewall gets borked. That "way" does not need to be accessible at all times, but you should be able to switch it on when needed.
Don't forget to test that these are actually working after you have sweet them up.

Whatever you decide on, don't forget that you don't have to do everything at once. Don't let it overload you. Learning new tech takes time.

Its the difference between running home things and having the money to run datacenter reliability.

I split my setup into storage vs processing. Can one physical box handle both? If the answer is yes then go for it. If all your running is low IO stuff and it's sipping cpu then one general purpose whitebox is a great start.

Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I've seen in this thread:

[Thread #867 for this sub, first seen 12th Jul 2024, 23:05] [FAQ] [Full list] [Contact] [Source code]