←back to thread

Ask HN: I have 24 core server with 1TB of DDR4 RAM, what should I run?

24 points pajeets | 1 comments | 17 Oct 24 17:45 UTC | HN request time: 0.202s | source

First time I'm doing colocation and as experiment I picked up a 1TB DDR4 RAM server with 24 cores (around 2.2ghz each)

What should I do now? What stuff should I run?

I have a hello world Flask app running but obviously its not enough to use the full potential.

I'm thinking of running KVM and selling a few VDS to friends or companies.

Also thought of running thousands of Selenium browser tests but I do this maybe once a year, not enough to fully utilize the server 24/7

Help! I might have gone overboard with server capacity, I will never have to pay for AWS again, I can literally run every single project, APIs, database I want and still have space left over.

1. freedomben ◴[17 Oct 24 20:35 UTC] No.41873492[source]▶
I would definitely run KVM, and set up virt-manager locally and connect to the server remotely over SSH. Whether you sell VPSs to friends or not, you can use it as a quick scratch pad for spinning up different os's for various testing purposes, both long lived and short.

I have a similar piece of hardware but 256 GB instead of a terabyte of RAM, and that is what I do. It has come in incredibly convenient to be able to spin up VMs as needed. I started creating different VMs for purposes. I would have normally just used the same host for, and have really enjoyed it.

I also run about a dozen personal services on there, such as audio bookshelf, archive box, jellyfin, navidrome, and more. Surprisingly, the archive box instance uses quite a bit of memory and CPU, so the box does get a fair amount of exercise. I have not looked very closely, but I believe archive box is using that memory and compute mainly for running headless Chrome. I have a browser extension installed that archives nearly every page I visit automatically, so especially during busy browsing times, I keep that thing running pretty hot.

In the past I set up a self-hosted openshift instance on it, spread across six VMs. I actually loved that and the only reason I'm no longer running it is because I broke it by messing around with risky and dangerous things that no sane person should ever do, and then did not want to dedicate the time to rebuild it. Someday I will recreate it again.

Whatever you decide to do with it, this is a really awesome problem to have!