Avoid rerunning no-op logical optimizer rules#22412
Closed
Dandandan wants to merge 4 commits into
Closed
Conversation
Contributor
Author
|
run benchmark sql_planner |
|
🤖 Criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing optimizer-skip-unchanged-noop-rules (2e0e199) to 0da8961 (merge-base) diff File an issue against this benchmark runner |
github-actions
Bot
added
optimizer
Contributor
Author
|
run benchmark sql_planner |
|
🤖 Criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
…anged-noop-rules # Conflicts: # datafusion/optimizer/src/push_down_filter.rs
|
🤖 Criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing optimizer-skip-unchanged-noop-rules (cfebeca) to 0da8961 (merge-base) diff File an issue against this benchmark runner |
- Move duplicated `transformed_if_changed` into `utils.rs` and use it from `decorrelate_predicate_subquery`, `eliminate_cross_join`, `optimize_unions`, and `push_down_join`. - `eliminate_duplicated_expr`: track change inline via length comparison and skip the `Aggregate::try_new`/`Sort` rebuild when no duplicates are present — avoids both the helper call and the schema-recompute. - `optimize_projections`: avoid cloning the whole `Window`/`TableScan` when nothing changes; OR `aggregate_changed` into the recursive result so the `Aggregate` branch reports `Transformed::yes` when FD pruning reduces group_expr/aggr_expr counts. - `optimizer.rs`: extract the `Transformed::yes`/`no` consistency check into `assert_transformed_matches_plan` so the loop stays readable. - `push_down_filter`: revert the simplify-predicate check to a length-based comparison (matches `simplify_predicates` actual semantics — it only changes content via merging, not reordering). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Contributor
Author
|
run benchmark sql_planner |
|
🤖 Criterion benchmark running (GKE) | trigger CPU Details (lscpu)Comparing optimizer-skip-unchanged-noop-rules (29096cf) to d318324 (merge-base) diff File an issue against this benchmark runner |
|
🤖 Criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
|
🤖 Criterion benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagebase (merge-base)
branch
File an issue against this benchmark runner |
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
2 participants
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
The logical optimizer currently reruns every rule on each pass. If a rule already returned
Transformed::noand no later rule has changed the plan since then, rerunning that rule does not add new information and adds planning cost.What changes are included in this PR?
This tracks a cheap plan version inside
Optimizer::optimize. The version increments whenever a rule reportstransformed = true. If a rule previously returnedTransformed::nofor the current plan version, the optimizer skips rerunning it until some rule changes the plan.Are these changes tested?
Existing tests
Are there any user-facing changes?
No