I have a spare 3070 GPU, as well as 16GB of Memory and my friend has a spare PSU, this part list has everything else I would need+everything I already have. Is there anything I should tweak or modify or will this build work, I plan to use it as a headless server.
Use case, I currently run a docker swarm cluster with two older Optiplexes and a raspberry Pi, like I said before, I have a spare PSU, GPU and Memory and would rather put it to work then sell it. I would like to add this new PC to my cluster and utilize it for my home services and also learning. The only items I would really be buying is the case, cpu and board. I would like to run some local AI models on this PC as well.
OK, so what you probably won’t get much out of would be load balancing knowledge, from your description the CPU far outpaces everything else you have running services today. To get a good handle on that sort of thing, its handy to have comparable hardware for each node.
But the CPU is more than enough for most general task services, so yeah that will do fine. In terms of the GPU, yes, that will work for AI tasks as far as I know, most of the hardware I’m using for that is work stuff I get my hands on, so I couldn’t tell you much about the performance of the 3070 specifically, and I doubt a 6000 Ada as a reference w9uld be helpful, so maybe others can chime in on that aspect.
Since its mostly for learning, yeah, go for it. If you want to run i5 24x7, I’d probably want to separate out some of that CPU from that PSU purely for power management/cost to run, but yes its more than adequate for most services you’d throw on there.
Most of the servers I’m running are using a CPU that came out about 5 years before that Ryzen, but they are also lower wattage systems. Since they dont need a ton of CPU at all times, this is more the ideal for continually running home services, but not the only way to do it.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !selfhosted@lemmy.world
A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don’t control.
Rules:
Be civil: we’re here to support and learn from one another. Insults won’t be tolerated. Flame wars are frowned upon.
No spam posting.
Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it’s not obvious why your post topic revolves around selfhosting, please include details to make it clear.
Don’t duplicate the full text of your blog or github here. Just post the link for folks to click.
Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).
No trolling.
Resources:
selfh.st Newsletter and index of selfhosted software and apps
See updated post!
OK, so what you probably won’t get much out of would be load balancing knowledge, from your description the CPU far outpaces everything else you have running services today. To get a good handle on that sort of thing, its handy to have comparable hardware for each node.
But the CPU is more than enough for most general task services, so yeah that will do fine. In terms of the GPU, yes, that will work for AI tasks as far as I know, most of the hardware I’m using for that is work stuff I get my hands on, so I couldn’t tell you much about the performance of the 3070 specifically, and I doubt a 6000 Ada as a reference w9uld be helpful, so maybe others can chime in on that aspect.
Since its mostly for learning, yeah, go for it. If you want to run i5 24x7, I’d probably want to separate out some of that CPU from that PSU purely for power management/cost to run, but yes its more than adequate for most services you’d throw on there.
Most of the servers I’m running are using a CPU that came out about 5 years before that Ryzen, but they are also lower wattage systems. Since they dont need a ton of CPU at all times, this is more the ideal for continually running home services, but not the only way to do it.
So build away and enjoy