Open source TPU-friendly ragged paged attention kernel. #26920
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
copybara-service
bot
force-pushed
the
test_732369481
branch
from
be91ed1 to
a82aec2
Compare
Key features: * ***Support mixed prefill and decode*** to increase throughput for inference. (eg., ***5x*** speedup compared to padded Muti-Queries Paged Attention implementation for llama-3-8b.) * ***No explicit `swapaxes`*** for `seq_len` and `num_head` in pre/post kernel. The kernel takes `num_head` in 2nd minor as it naturally was. We fold swapaxes to strided load/store in the kernel and apply transpose on the fly. * ***No GMM (Grouped Matmul) Metadata required!*** We calculate the metadata on the fly in the kernel. This can speed up ***10%***! * ***Increase MXU utilization 8x in GQA*** by grouping shared q heads for MXU in decode. * ***Minimize recompilation:*** The only factors can cause recompilation are model specs, `max_num_batched_tokens` and `max_num_seqs` in the setting of mixed engine. PiperOrigin-RevId: 734269519
copybara-service
bot
force-pushed
the
test_732369481
branch
from
a82aec2 to
4b49c03
Compare
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Outdated suggestions cannot be applied.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.
Open source TPU-friendly ragged paged attention kernel.
Key features:
swapaxesforseq_lenandnum_headin pre/post kernel. The kernel takesnum_headin 2nd minor as it naturally was. We fold swapaxes to strided load/store in the kernel and apply transpose on the fly.max_num_batched_tokensandmax_num_seqsin the setting of mixed engine.