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

Improve RAM performance and organization of GPU2 work #1426

Closed
cryptonemo opened this issue Mar 5, 2021 · 2 comments
Closed

Improve RAM performance and organization of GPU2 work #1426

cryptonemo opened this issue Mar 5, 2021 · 2 comments

Comments

@cryptonemo
Copy link
Collaborator

Description

There are improvements noted in the PR #1422

Also, 32GiB benchmarking is required to determine the performance difference of gpu and gpu2 before moving further. See small GPU example graph in PR mentioned above. RAM usage and wall clock time considerations are also important to measure and optimize.

Acceptance criteria

Investigate if RAM usage can be improved with mmap or unnecessary threading can be simplified as noted (i.e. removal of unnecessary spawn).

Risks + pitfalls

Where to begin

#1422
@cryptonemo
Copy link
Collaborator Author

The first part of this has been resolved by #1427

Next up is experimenting with potentially improved performance.

@cryptonemo cryptonemo mentioned this issue Mar 18, 2021
Merged
@cryptonemo
Copy link
Collaborator Author

Closing due to #1436 for the initial investigation

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
1 participant
@cryptonemo