Repository navigation
feat: shuffle skew detection + bloom filter pre-filtering - #5
Merged
Merged
Conversation
…st joins Shuffle skew detection: - detect_skew/3 inspects bucket sizes after hash-partitioning - Heavy buckets (>5x mean) emit [:dux, :distributed, :shuffle, :skew_detected] telemetry with bucket IDs and sizes - New :over_partition_factor option for user tuning (default 4) Bloom filter pre-filtering: - Before broadcasting full right table, extract and broadcast distinct join keys as a separate table - Workers pre-filter their left partition with IN (SELECT key FROM keys_table) - DuckDB optimizes this as a semi-join internally - Only for :inner/:left joins (anti/semi need unfiltered left) - Skip if right has >10K distinct keys (filter wouldn't help) - Keys table cleaned up with other broadcast tables Fixes: - Multi-column COUNT DISTINCT: use subquery instead of COUNT(DISTINCT a, b) - Anti join: skip bloom filter (would eliminate rows we want to keep) 505 tests (43 doctests + 11 properties + 451 tests), 0 failures. 3x consecutive runs confirmed stable. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds two new benchmark sections: shuffle join (100K×100K) and broadcast join with bloom filter pre-filtering (100K×20). Fixes the version column in history.csv to use SHA-based identifiers instead of non-existent tags. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
[:dux, :distributed, :shuffle, :skew_detected]telemetry when any bucket exceeds 5× the mean. Adds configurable:over_partition_factoroption (default 4).IN (SELECT ...)before the main join, reducing bandwidth for selective joins. Only applies to:innerand:leftjoins (skipped for anti/right joins where filtering would drop needed rows). Skipped when distinct key count > 10,000.history.csvto use SHA-based identifiers.Test plan
🤖 Generated with Claude Code