Optimizations for Pipeline Parallel Serving #11702
Merged
Conversation
This file contains 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
qiyuangong
reviewed
Aug 1, 2024
| @@ -179,6 +180,11 @@ def pipeline_parallel(model, pipeline_parallel_stages, torch_dtype=torch.float32 | |||
| layer_start = slice_size * local_rank | |||
| layer_end = layer_start + min(slice_size, num_layers - layer_start) | |||
|
|
|||
| # if local_rank == 0: | |||
There was a problem hiding this comment.
Remove these lines if they are not necessary
qiyuangong
reviewed
Aug 1, 2024
| @@ -738,14 +744,73 @@ def clear_batch(self, cur_id): | |||
| self.is_finish.pop(cur_id, None) | |||
| self.partial_output_dict.pop(cur_id, None) | |||
|
|
|||
| async def finish_stream_output(self, cur_id): | |||
qiyuangong
reviewed
Aug 1, 2024
| torch.xpu.synchronize(self.device) | ||
| self.send_buff.wait() | ||
| if output is not None: | ||
| self.send_buff = dist.isend(output, dst=self.next_rank) |
There was a problem hiding this comment.
https://pytorch.org/docs/stable/distributed.html#torch.distributed.isend
Use async isend to replace sync send.
qiyuangong
reviewed
Aug 1, 2024
| # remain = 0 | ||
| # self.is_finish[request_id] = True | ||
|
|
||
| text = cur_text[cached_index] |
qiyuangong
changed the title
qiyuangong
reviewed
Aug 2, 2024
|
|
||
| text = tokenizer.decode(self.token_cache[request_id]) | ||
|
|
||
| if text.endswith("\n"): |
There was a problem hiding this comment.
This block is text-related. Maybe we can move it into a function,
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Outdated suggestions cannot be applied.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.
3. Summary of the change
Optimize
stream_output()methods in Pipeline Parallel Serving.4. How to test?
1234). And paste your action link here once it has been successfully finished.https://github.com/intel-analytics/ipex-llm-workflow/actions/runs/10209347945