Manya0407 commented on code in PR #6746:
URL: https://github.com/apache/hive/pull/6746#discussion_r4121682849
##########
ql/src/test/results/clientpositive/cbo_rp_auto_join1.q.out:
##########
@@ -853,40 +856,149 @@ STAGE PLANS:
expressions: key (type: int)
outputColumnNames: key
Statistics: Num rows: 7 Data size: 28 Basic stats: COMPLETE
Column stats: COMPLETE
- Sorted Merge Bucket Map Join Operator
Review Comment:
@zabetak I looked at cbo_rp_auto_join1. With
hive.cbo.returnpath.hiveop=true and hive.transpose.aggr.join=true, agg-join
transpose in CBO was producing plans with bucketGroup + reduce-side join, so
Sorted Merge Bucket Map Join disappeared (vs master). Updated selectivity
likely made transpose win on cost more often; the underlying issue is that
transpose + return path yields a shape the bucket/SMB rules don’t convert.
Fix: skip Calcite agg-join transpose when CBO return path is enabled
(CalcitePlanner.java), regen cbo_rp_auto_join1.q.out. Locally TestCliDriver /
cbo_rp_auto_join1.q passes and SMB is back.I’ll push this in the next commit
unless you’d prefer fixing it via selectivity instead...
--
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]
