perf: Vectorized condition evaluation for CASE WHEN with many branches [WIP] #19589
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
This adds a new optimization path for searched CASE WHEN expressions (CASE WHEN condition THEN result) with 3 or more branches. Instead of evaluating conditions sequentially on progressively shrinking batches, this approach: 1. Evaluates all conditions upfront on the full batch 2. Builds a branch_index array indicating which branch matched each row 3. Filters once per branch and evaluates THEN expressions This provides better performance due to: - Better cache locality (conditions evaluated on same data) - Simpler filter predicates (integer equality vs boolean expressions) - No progressive batch shrinking overhead **Important**: This changes short-circuit semantics for CONDITIONS (not THEN expressions). All conditions are evaluated even for rows where an earlier condition matched. This is safe for simple comparisons but may cause issues if conditions can error (e.g., division by zero in a condition). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.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.
Which issue does this PR close?
Rationale for this change
I ran some microbenchmarks comparing DataFusion and DuckDB (see apache/datafusion-benchmarks#28) and found that
CASE WHENexpressions were much slower in DataFusion, so I asked Claude to make it go faster.What changes are included in this PR?
This adds a new optimization path for searched CASE WHEN expressions (CASE WHEN condition THEN result) with 3 or more branches.
Instead of evaluating conditions sequentially on progressively shrinking batches, this approach:
This provides better performance due to:
Important: This changes short-circuit semantics for CONDITIONS (not THEN expressions). All conditions are evaluated even for rows where an earlier condition matched. This is safe for simple comparisons but may cause issues if conditions can error (e.g., division by zero in a condition).
🤖 Generated with Claude Code
Are these changes tested?
Existing tests
Are there any user-facing changes?
No