Optimized LL implementation - #834
Merged
Merged
Conversation
Quantize BF16 dispatch payloads to FP8 E4M3 with format-defined block scales while preserving BF16 expert outputs for combine. Clean up the sender structure, payload metadata, vector conversions, Python API, and multi-rank coverage. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Define focused read-only review guidance for CUDA/HIP synchronization, memory ordering, communication correctness, data types, and GPU performance. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Refresh the Python and C++ benchmark paths for BF16 and FP8 dispatch, current MoERuntime signatures, active kernel sources, portable CUPTI discovery, realistic routing, and safe unified reporting. Remove the merged change to the inactive legacy implementation. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Resolve the C++ benchmark conflict by combining NCCL-EP-style random top-k routing and masked selections with the current BF16/FP8 MoERuntime API and format-aware byte accounting. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Resolve the FP8 conversion conflict by using direct float2-to-E4M3 conversion on all CUDA architectures while preserving the MSCCL++ packed vector storage path. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Use Bootstrap node and NVML fabric-domain discovery for optimized LL, collectively gate cross-host fabric payload and semaphore support, remove obsolete topology env defaults, and refresh the EP documentation and performance guidance. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
Use fixed symmetric receive buffers and direct peer mappings, simplify kernel synchronization and workspace organization, and move LL buffer sizing into C++. Remove legacy PortChannel sources and unnecessary Python RDMA APIs. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: efbacae6-f679-430b-bc16-b45ae162fc76
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
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
1 participant
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.
No description provided.