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

GPUDirect for NVMe Offload #3063

Closed
srikanthmalla opened this issue Mar 20, 2023 · 2 comments
Closed

GPUDirect for NVMe Offload #3063

srikanthmalla opened this issue Mar 20, 2023 · 2 comments

Comments

@srikanthmalla
Copy link

Hi,
Does DeepSpeed use GPUDirect for cuda>=11.4 (because it is available by default), more specifically for NVMe offloading? If so, how to verify if it is used or not?

Thank you!
@srikanthmalla srikanthmalla changed the title GPUDirect GPUDirect for NVMe Offload Mar 20, 2023
@tjruwase
Copy link
Contributor

@srikanthmalla, we plan to add this support soon. Thanks!

@tjruwase
Copy link
Contributor

@srikanthmalla, GDS support is now integrated into DeepSpeed. We have enabled for NVMe offload of ZeRO-Inference.

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
@tjruwase @srikanthmalla