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

jaxdecomp proto #21

Open
wants to merge 29 commits into
base: main
Choose a base branch
from
Open

jaxdecomp proto #21

wants to merge 29 commits into from

Conversation

ASKabalan
Copy link
Collaborator

I am going to continue here the distributed PM

@ASKabalan ASKabalan marked this pull request as draft July 9, 2024 22:27
EiffL
EiffL reviewed Jul 19, 2024
View reviewed changes
scripts/distributed_pm.py Outdated Show resolved Hide resolved
ASKabalan
ASKabalan commented Jul 19, 2024
View reviewed changes
scripts/particle_mesh.slurm
@@ -0,0 +1,167 @@
#!/bin/bash
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This file should not stay in the repo in my opinion

@ASKabalan ASKabalan marked this pull request as ready for review August 2, 2024 22:26
@ASKabalan
Copy link
Collaborator Author

ASKabalan commented Aug 2, 2024 •
edited
Loading

@EiffL
updated for the new branch in jaxdecomp with the single transposition SLAB
I have verified visually and against PMWD
I am now benchmarking LPT, LeapFrog against pmwd on GPUs from 1 to 128 and from sizes from 256 to 4096
I can already see that the XY slab is on average 36% faster than pencils
What would be a good way to test jaxPM?
I can make some unit tests to make sure that functions like cic_read and cic_paint are consistent
But in general, should I add tests comparing fields with pmwd and fastPM?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@EiffL EiffL EiffL left review comments

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@ASKabalan @EiffL