Repository navigation
[PD] Opt-in Core transfer path for NIXL — review MVP - #1
Draft
samnordmann wants to merge 2 commits into
Draft
samnordmann wants to merge 2 commits into
samnordmann wants to merge 2 commits into
Conversation
Snapshot of the current endpoint-transfer prototype for review in the author's fork. Native GPU performance parity and full end-to-end framework validation remain open; this is not an upstream-ready change. Assisted-by: OpenAI Codex Signed-off-by: Samuel Nordmann <snordmann@nvidia.com>
This was referenced Sep 17, 2026
Restore shared lifecycle, wire and staging code to the review base. Wrap the existing NIXL agent only when explicitly enabled; retain the native default. Assisted-by: OpenAI Codex Signed-off-by: Samuel Nordmann <snordmann@nvidia.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation / modifications
An opt-in Core transfer experiment using
SGLANG_DISAGGREGATION_NIXL_USE_TORCH_TRANSFER=1. Wrap the existing NIXL agent; default remains unchanged. Preserve bootstrap, staging, CUDA ordering, wire messages and allocation/lifecycle policy.Requires Core MVP and NIXL provider/shim. The prior shared-protocol/Rust/Mooncake changes have been removed.
Accuracy / speed tests
Modified Python files passed static checks. The existing
test_nixl_backend_basic.pysuite could not collect locally: missingorjson. Shared shim contract tests passed (4 CPU tests with mocked native calls); this is not an SGLang runtime result.No serving/GPU or speed validation. Success-path review prototype only; active abort/recovery is not supported. AI-assisted fork draft; human review and end-to-end validation pending.