You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
I use docker installation method. Since I have multiple systems that the number of GPUs are differ, so bind mount every GPU to every contains seem not good enough.
Describe the solution you'd like
I would like to be able to run a same docker-compose file on every system.
Describe alternatives you've considered
I tried to rebuild your Dockerfile, with from nvidia/cuda:11.6.2-base-ubuntu20.04 instead of ubuntu:22.04, as following https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html.
it's not work since the Dockerfile provided is not work. If you please make the Dockerfile actually works, I will try to make a pull request.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
I use docker installation method. Since I have multiple systems that the number of GPUs are differ, so bind mount every GPU to every contains seem not good enough.
Describe the solution you'd like
I would like to be able to run a same
docker-compose
file on every system.Describe alternatives you've considered
I tried to rebuild your Dockerfile, with from
nvidia/cuda:11.6.2-base-ubuntu20.04
instead ofubuntu:22.04
, as following https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html.it's not work since the Dockerfile provided is not work. If you please make the Dockerfile actually works, I will try to make a pull request.
The text was updated successfully, but these errors were encountered: