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.

Sign up for GitHub

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

Jump to bottom

Parallel GPU use #97

Open
Jay-Beeb opened this issue Dec 9, 2024 · 1 comment
Open

Parallel GPU use #97

Jay-Beeb opened this issue Dec 9, 2024 · 1 comment

Comments

@Jay-Beeb
Copy link

Jay-Beeb commented Dec 9, 2024

Hi!
Thanks for sharing this powerful tool!
I'm trying to predict complexes with multiple sequences and run out of memory fast. I noticed that only one of 4 GPUs is used during the prediction. It has 94GB of RAM and I'm able to predict complexes with 4 proteins (2000AA total) but not more. Is this normal memory usage and is there a way to use more GPUS in parallel? I tried --devices, but it seems to only spread predictions for different entities to other gpus.
Thanks in advance!

Jay
@jwohlwend
Copy link
Owner

We do not support sharding across GPU's at this time but you should be able to fold 2000 AA's with an 80G GPU (maybe even 48G)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@jwohlwend @Jay-Beeb