perf: skip empty memtable during query scoring#252
Merged
Conversation
When the memtable has no postings (total_postings == 0), return NULL from tp_memtable_source_create to skip the entire memtable scoring path. This avoids dshash_attach (x2), hash_create for doc accumulation, per-term dshash_find lookups, and dshash_detach (x2) -- all of which were pure overhead after CREATE INDEX since the memtable is empty. Profiling showed tp_memtable_search at 16% of CPU on 138M rows where every memtable lookup returned nothing.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
tp_memtable_source_createwhenmemtable->total_postings == 0Motivation
After
CREATE INDEX, the memtable is empty -- all data lives in segments. But every query still attached to 2 dshash tables, created a doc accumulation hash table, and did per-term hash lookups that always returned NULL. Profiling on 138M MS-MARCO v2 showed this at 16% of CPU.Test plan