Port CUDA SSD streaming support#349
Open
linuxbest wants to merge 1 commit into
Open
Conversation
Add CUDA support for SSD streaming async selected loads and shared overlap decode paths, while preserving the existing Metal streaming path and ds4_lx session cancellation API. Also update SSD streaming help text so it is described as a GPU graph backend feature instead of Metal-only. Verification: make ds4; make ds4-server ds4-bench ds4-eval ds4-agent; make cuda-regression; make ds4 ds4-bench; git diff --check. Quick cold-cache benchmark on NVIDIA GB10 with 8GB expert cache: prefill 2048 tokens at ~30-32 tok/s, decode 32 tokens at ~2.1 tok/s.
Author
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add CUDA support for SSD streaming async selected loads and shared overlap decode paths, while preserving the existing Metal streaming path and ds4_lx session cancellation API.
Also update SSD streaming help text so it is described as a GPU graph backend feature instead of Metal-only.
Verification: make ds4; make ds4-server ds4-bench ds4-eval ds4-agent; make cuda-regression; make ds4 ds4-bench; git diff --check.
Quick cold-cache benchmark on NVIDIA GB10 with 8GB expert cache: prefill 2048 tokens at ~30-32 tok/s, decode 32 tokens at ~2.1 tok/s.