←back to thread

I regret building this $3000 Pi AI cluster

(www.jeffgeerling.com)
468 points speckx | 1 comments | 19 Sep 25 14:28 UTC | HN request time: 0.203s | source
Show context
Aurornis ◴[19 Sep 25 14:52 UTC] No.45302320[source]▶
I thought the conclusion should have been obvious: A cluster of Raspberry Pi units is an expensive nerd indulgence for fun, not an actual pathway to high performance compute. I don’t know if anyone building a Pi cluster actually goes into it thinking it’s going to be a cost effective endeavor, do they? Maybe this is just YouTube-style headline writing spilling over to the blog for the clicks.

If your goal is to play with or learn on a cluster of Linux machines, the cost effective way to do it is to buy a desktop consumer CPU, install a hypervisor, and create a lot of VMs. It’s not as satisfying as plugging cables into different Raspberry Pi units and connecting them all together if that’s your thing, but once you’re in the terminal the desktop CPU, RAM, and flexibility of the system will be appreciated.

replies(11): >>45302356 #>>45302424 #>>45302433 #>>45302531 #>>45302676 #>>45302770 #>>45303057 #>>45303061 #>>45303424 #>>45304502 #>>45304568 #
kolbe ◴[19 Sep 25 16:20 UTC] No.45303424[source]▶
The author, Jeff Geerling, is a very intelligent person. He has more experience with using niche hardware than almost anyone on earth. If he does something, there's usually a good a priori rationale for it.
replies(3): >>45303517 #>>45303850 #>>45304400 #
buildbot ◴[19 Sep 25 16:28 UTC] No.45303517[source]▶
Jeff is a good person/blogger and does interesting projects but more experience with niche hardware than literally anyone is a stretch.

Like what about the people who maintain the alpha/sparc/parisc linux kernels? Or the designers behind idk tilera or tenstorrent hardware.

replies(3): >>45303646 #>>45303707 #>>45303825 #
1. geerlingguy ◴[19 Sep 25 16:44 UTC] No.45303707[source]▶
I was just at VCF Midwest this past weekend, and I can assure you I am on some of the lower echelons of people who know about niche hardware.

I do get to see and play with a lot of interesting systems, but for most of them, I only get to go just under surface-level. It's a lot different seeing someone who's reverse engineered every aspect of an IBM PC110, or someone who's restored an entire old mainframe that was in storage for years... or the group of people who built an entire functional telephone exchange with equipment spread over 50 years (including a cell network, a billing system, etc.).