impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bharath Vissapragada (Code Review)" <ger...@cloudera.org>
Subject [Impala-ASF-CR] IMPALA-5612: join inversion should factor in parallelism
Date Mon, 10 Jul 2017 05:57:04 GMT
Bharath Vissapragada has posted comments on this change.

Change subject: IMPALA-5612: join inversion should factor in parallelism
......................................................................


Patch Set 3: Code-Review+1

(4 comments)

http://gerrit.cloudera.org:8080/#/c/7351/3//COMMIT_MSG
Commit Message:

Line 19: 
May be add a note on testing and the perf difference you see on TPCH/DS and if any specific
queries regressed, just so its documented somewhere in the git history.


http://gerrit.cloudera.org:8080/#/c/7351/2/fe/src/main/java/org/apache/impala/planner/Planner.java
File fe/src/main/java/org/apache/impala/planner/Planner.java:

Line 386:    *    Do not invert if relevant stats are missing.
> Updated #3 to include a pointer to isInvertedJoinCheaper(). There were some
Thanks, looks better now.


PS2, Line 424: invertedJoinIsCheaper
> Done
Missed this?


PS2, Line 459: itrary unit of time, is then:
> It's really execution time, with an arbitrary unit.
Thanks for explaining. I now see why you multiply 2 * rhsCard by the log. Makes sense to me.


-- 
To view, visit http://gerrit.cloudera.org:8080/7351
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Icacea4565ce25ef15aaab014684c9440dd501d4e
Gerrit-PatchSet: 3
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-Reviewer: Alex Behm <alex.behm@cloudera.com>
Gerrit-Reviewer: Bharath Vissapragada <bharathv@cloudera.com>
Gerrit-Reviewer: Mostafa Mokhtar <mmokhtar@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <tarmstrong@cloudera.com>
Gerrit-HasComments: Yes

Mime
View raw message