2

I am a Data Scientist currently working on non-AI/ML projects, but I would like to set up a desk system to start learning AI/ML. I have an iMac and a MacBook Pro, they are both Intel-based and are excellent computers, but very limited on GPU. I am considering either attaching an eGPU, or building a home server with a Linux box that I will use for processing.

I understand that Mac computers do not use CUDA, so I think I will be limited to AMD GPUs if I use an eGPU enclosure. If I set up a home server running Linux, I could use Nvidia.

I am also aware that an eGPU is bandwidth-limited by the TB3 connection, vice PCIe.

Should I pursue a eGPU enclosure, or should I just build a Linux box with two GPUs on PCIe? Money isn't too much of an issue.

Giacomo1968
  • 231
  • 1
  • 11
Seth L.
  • 31
  • 3
  • 3
    eGPUs are only supported on Intel Macs; are you using an Apple Silicon or an Intel Mac? In terms of performance, you'll definitely be better off doing a dedicated Linux box with Nvidia GPUs; if money isn't an issue, definitely do that. – JMY1000 Feb 01 '24 at 01:20
  • @JMY1000, Intel Macs. Thanks for the advice. I wasn't aware that eGPUs don't work on Apple Silicon. Although my Macs can support eGPU as you say, it seems like a Linux server would be the better option at least in performance--that's what I had previously assumed, but I'd never seen any actual statistics. I'd love to see how it would compare to using an eGPU in terms of benchmarks.

    My command line and Linux knowledge are decent, but looks like I'll be needing to improve them...

     – Seth L. Feb 06 '24 at 02:42

0 Answers0