You are viewing a plain text version of this content. The canonical link for it is here.

Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2021/06/01 12:55:48 UTC

[GitHub] [spark] ulysses-you commented on pull request #32683: [SPARK-35540][SQL] Make config maxShuffledHashJoinLocalMapThreshold fallback to advisoryPartitionSizeInBytes

ulysses-you commented on pull request #32683:
URL: https://github.com/apache/spark/pull/32683#issuecomment-852101332


   Here list some queries with obvious performance differences. Note that, seems query `q72` has a perf regression .
   
   tpcds sf100 perf number(seconds) with two value of `spark.sql.adaptive.maxShuffledHashJoinLocalMapThreshold`:
   
   query|64mb|0|diff
   -|-|-|-
   q2|72.224|109.094|0.337965424
   q11|87.811|106.552|0.175885952
   q14a|269.834|298.432|0.095827525
   q14b|200.496|212.351|0.05582738
   q32|24.37|27.066|0.099608365
   q37|31.595|36.413|0.132315382
   q57|40.86|46.324|0.117951818
   q69|42.891|52.873|0.188792011
   q72|124.234|98.2|/
   


-- 
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org