I have a laptop and a desktop with a decent GPU connected via ssh, both running ubuntu 18.04. I would like to send GPU tasks, to the server as an EGPU set up. Both are running cuda, but my laptop runs out of GPU memory, so i would like to send out the GPU intensive tasks. I dont know an efficient way to do that, if anyone knows how to set this up I would surely be interested. I will be running darknet, and other programs written in C/C++ using openCV

The server GPU from lspci:

controller: NVIDIA Corporation GP106 [GeForce GTX 1060 6GB] (rev a1)
  • 1
    Your problem is hardware related; not software. Do you have thunderbolt ports on both ends? – Rinzwind Jan 12 at 22:46
  • no I dont have thunderbolt, but do I really need that? I just want to send my video stream to the server, have it process the matrix, then send back the processed stream. – j0h Jan 13 at 1:40

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.