Open
Conversation
There was a problem hiding this comment.
Pull Request Overview
This PR improves performance of repartitioned multi-GPU matrix values and vector value update operations by optimizing communication patterns and adding device communication initialization.
- Refactors all-to-all communication to use MPI operations (MPI_Igatherv, MPI_Iscatterv) for better performance
- Adds symmetric matrix optimization to skip redundant data transfers
- Introduces device communication initialization and repartitioned communicator support
Reviewed Changes
Copilot reviewed 9 out of 11 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| test/unit/MatrixWrapper/Distributed.cpp | Adds device communication initialization calls to test setup |
| src/StoppingCriterion.cpp | Adds timing measurements for performance analysis |
| src/MatrixWrapper/Distributed.cpp | Implements optimized MPI communication and symmetric matrix handling |
| src/CommunicationPattern.cpp | Adds new function for computing repartitioned all-to-all patterns |
| include/OGL/lduLduBase.hpp | Adds device communication initialization and timing code |
| include/OGL/StoppingCriterion.hpp | Adds frequency mode configuration for stopping criterion optimization |
| include/OGL/DevicePersistent/Vector.hpp | Replaces communication function with direct MPI calls for copy-back operation |
| include/OGL/DevicePersistent/ExecutorHandler.hpp | Adds rank management utilities and device communication setup |
| include/OGL/CommunicationPattern.hpp | Declares new repartitioned communication pattern function |
Comments suppressed due to low confidence (1)
src/MatrixWrapper/Distributed.cpp:1
- Duplicate assignment to
recv_offsets.back(). The first assignment on line 203 setssend_offsets.back()but the code incorrectly assigns torecv_offsets.back()twice. Line 203 should assign tosend_offsets.back().
// SPDX-FileCopyrightText: 2024 OGL authors
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
Comment on lines
+345
to
+346
| std::string("\n\tRetrieve results bandwidth "); // + | ||
| std::to_string(bandwidth_copy_back) + std::string(" [GByte/s]"); |
There was a problem hiding this comment.
The string concatenation is broken. Line 345 ends with a comment // + and line 346 is a separate statement that isn't assigned to anything. The + operator should be on line 345 to concatenate with line 346.
Suggested change
| std::string("\n\tRetrieve results bandwidth "); // + | |
| std::to_string(bandwidth_copy_back) + std::string(" [GByte/s]"); | |
| std::string("\n\tRetrieve results bandwidth ") + | |
| std::to_string(bandwidth_copy_back) + std::string(" [GByte/s]"); |
323dc85 to
11a7a16
Compare
82efe8e to
28ee971
Compare
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
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 improves repartitioned multi gpu matrix values and vector value update performance.