Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

dockerfiles for gpu with cudnn are missing #606

Open
laszewsk opened this issue May 30, 2024 · 1 comment
Open

dockerfiles for gpu with cudnn are missing #606

laszewsk opened this issue May 30, 2024 · 1 comment
Labels
type: feature Issues that include feature request or feature idea

Comments

@laszewsk
Copy link

Description

Dockerfiles for smartsim using gpus with cudnn are missing

Justification

everyone that wants to use smartsim with gpus and docker

Implementation Strategy

the dockerfiles in https://github.com/CrayLabs/SmartSim/tree/develop/docker only contain device=cpu, but if you change it to gpu, the cudnn library is missing so smartsim does not work.

I suggest adding a new dir gpu in which a proper gpu based Dockerfile for this is placed, one coupd distingusih
gpu-nvidia
gpu-rocm
if needed. I only need nvidia

@laszewsk laszewsk added the type: feature Issues that include feature request or feature idea label May 30, 2024
@ashao
Copy link
Member

ashao commented Jun 4, 2024

Until recently, Nvidia did not have an 'official' pathway to download cuDNN without registering with their developer zone. This now seems possible, so we could explore the possibility of updating our tutorial Docker container to support this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
type: feature Issues that include feature request or feature idea
Projects
None yet
Development

No branches or pull requests

2 participants