feiniaofeiafei opened a new pull request, #66117:
URL: https://github.com/apache/doris/pull/66117
### What problem does this PR solve?
Issue Number: close #xxx
Related PR: #xxx
Problem Summary:
Adjust the “large join” threshold by increasing the build-side row-count
limit from 400K to 1M.
Allow aggregation to continue pushing down to the current join branch when
the opposite subtree is large or outputs String/Character types, so that data
participating in the join can be reduced earlier.
Tighten aggregation benefit evaluation: treat Group By keys with NDV >
rowCount / 10 as having extremely low aggregation benefit; for multi-column
Group By, skip aggregation pushdown if any key falls into this category,
avoiding low-benefit pushdown caused by high cardinality.
Take small broadcast join into account and propagate a “bottom small
broadcast join” marker in the pushdown context and apply stricter checks on the
probe side: reject pushdown when Group By column statistics are unknown, or
when any Group By key is estimated to reduce rows by less than 1000×.
Extend aggregation pushdown traversal to support passing through
LogicalAggregate and LogicalIntersect nodes.
### Release note
None
### Check List (For Author)
- Test <!-- At least one of them must be included. -->
- [ ] Regression test
- [ ] Unit Test
- [ ] Manual test (add detailed scripts or steps below)
- [ ] No need to test or manual test. Explain why:
- [ ] This is a refactor/code format and no logic has been changed.
- [ ] Previous test can cover this change.
- [ ] No code files have been changed.
- [ ] Other reason <!-- Add your reason? -->
- Behavior changed:
- [ ] No.
- [ ] Yes. <!-- Explain the behavior change -->
- Does this need documentation?
- [ ] No.
- [ ] Yes. <!-- Add document PR link here. eg:
https://github.com/apache/doris-website/pull/1214 -->
### Check List (For Reviewer who merge this PR)
- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label <!-- Add branch pick label that this PR should
merge into -->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
