Launch embedding server earlier #176
Open
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
- Reproduced the slow search performance issue (15-30s vs expected ~2s) - Identified root cause: default complexity=64 is too high for fast search - Created test script demonstrating performance with different complexity values - Test results show complexity=16-32 achieves ~2s search time (matching paper) - Added comprehensive analysis document with solutions and recommendations Key findings: - Default complexity=64 results in ~36s search time - Reducing complexity to 16-32 achieves ~2s search time - beam_width parameter is mainly for DiskANN, not HNSW - Paper likely used smaller embedding model (~100M) and lower complexity Solutions provided: 1. Reduce complexity parameter to 16-32 for faster search 2. Consider DiskANN backend for better performance on large datasets 3. Use smaller embedding model if speed is critical
- Test script to reproduce slow search performance issue - Generates ~90K chunks (~180MB) similar to user's dataset - Tests search performance with different complexity values (8, 16, 32, 64) - Demonstrates that complexity=16-32 achieves ~2s search time - Validates the performance analysis findings
Collaborator
Author
|
@Ai-yang-dev Can you take a look here? |
Owner
|
@andylizf also add the logic to keep the embedding server alive, and add command to kill that |
Fine. Thanks for Sharing. |
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Outdated suggestions cannot be applied.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.
Supersedes #165.
recomputeparameter fromSearcher.searchtoSearcher.__init__in this PR.manual_tokenizecan be used to fasten the embedding generation, and thus fasten the search process.