tutorials for LSM and MDD using NCCL #156
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mrava87
reviewed
Jul 9, 2025
mrava87
reviewed
Jul 9, 2025
|
@tharittk nearly ready, just left two comments above as I think you have a couple of typos that may not make things render properly? |
mrava87
approved these changes
Jul 10, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR is for the documentation of Multi-Dimensional Deconvolution (MDD) and Least-Squares Migration (LSM) under NCCL communication.
It was tested on 3 GPUs and ran successfully.
For quick stats on 3 GPUs (UIUC Delta)
LSM (iterations = 100)
CuPy + NCCL Total time (s) = 1.96
CuPy + MPI. Total time (s) = 1.94
NumPy + MPI Total time (s) = 1.19
MDD (iterations = 50)
CuPy + NCCL Total time (s) = 1.62
CuPy + MPI Total time (s) = 2.57
NumPy + MPI Total time (s) = 6.83
The LSM result is not yet as expected. It could be that the input size is too small for the GPU computations.
Further tests will be carried out.