[
https://issues.apache.org/jira/browse/SPARK-3711?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14152878#comment-14152878
]
Yash Datta commented on SPARK-3711:
-----------------------------------
On a 2 node setup
each machine config: 24 core machine , (96 GB )
invoking spark-sql with :
./bin/spark-sql --executor-memory 16G --driver-memory 8G --master <url>
executing a filter query on a parquet table having 47750544 rows , with ~1000
filters:
(the selected column was unique for each row)
select * from <table> where <column> in (A1,A2....A1000);
Time taken on spark-1.1 (after multiple runs) :
~90 seconds
after the patch :
~7 seconds
> Optimize where in clause filter queries
> ---------------------------------------
>
> Key: SPARK-3711
> URL: https://issues.apache.org/jira/browse/SPARK-3711
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 1.1.0
> Reporter: Yash Datta
> Priority: Minor
> Fix For: 1.1.1
>
>
> The In case class is replaced by a InSet class in case all the filters are
> literals, which uses a hashset instead of Sequence, thereby giving
> significant performance improvement. Maximum improvement should be visible in
> case small percentage of large data matches the filter list
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
