fix(vector): filter INVALID_OFFSET neighbors in CSR length and numRels#27
Merged
Conversation
populateCSRLengths and getNumRelsInGraph used graph.getCSRLength, which counts all occupied slots including ones later marked INVALID_OFFSET, while writeToTable skipped those neighbors. This left the rel table with fewer rows than its CSR length/offsets indicated, producing holes/stale values in the chunked group. Add getNumValidNeighbors helper that mirrors writeToTable's filtering and use it in both populateCSRLengths and getNumRelsInGraph so the CSR metadata stays consistent with the rel rows actually written. Add HNSWRelBatchInsertTest covering index creation + query across multiple nodes/groups.
adsharma
force-pushed
the
more_fixes_and_tests
branch
from
41e21d8 to
b5b982e
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Related: #23
populateCSRLengths and getNumRelsInGraph used graph.getCSRLength, which counts all occupied slots including ones later marked INVALID_OFFSET, while writeToTable skipped those neighbors. This left the rel table with fewer rows than its CSR length/offsets indicated, producing holes/stale values in the chunked group.
Add getNumValidNeighbors helper that mirrors writeToTable's filtering and use it in both populateCSRLengths and getNumRelsInGraph so the CSR metadata stays consistent with the rel rows actually written.
Add HNSWRelBatchInsertTest covering index creation + query across multiple nodes/groups.