spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sathiya Kumar (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (SPARK-22181) ReplaceExceptWithNotFilter if one or both of the datasets are fully derived out of Filters from a same parent
Date Sun, 01 Oct 2017 16:49:00 GMT

     [ https://issues.apache.org/jira/browse/SPARK-22181?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sathiya Kumar updated SPARK-22181:
----------------------------------
    Description: 
While applying Except operator between two datasets, if one or both of the datasets are purely
transformed using filter operations, then instead of rewriting the Except operator using expensive
join operation, we can rewrite it using filter operation by flipping the filter condition
of the right node.

![Case-1](https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case1.png)

![Case-2](https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case2.png)


  was:
While applying Except operator between two datasets, if one or both of the datasets are purely
transformed using filter operations, then instead of rewriting the Except operator using expensive
join operation, we can rewrite it using filter operation by flipping the filter condition
of the right node.

!https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case1.png!

!https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case2.png!


> ReplaceExceptWithNotFilter if one or both of the datasets are fully derived out of Filters
from a same parent
> -------------------------------------------------------------------------------------------------------------
>
>                 Key: SPARK-22181
>                 URL: https://issues.apache.org/jira/browse/SPARK-22181
>             Project: Spark
>          Issue Type: New Feature
>          Components: Optimizer, SQL
>    Affects Versions: 2.1.1, 2.2.0
>            Reporter: Sathiya Kumar
>            Priority: Minor
>
> While applying Except operator between two datasets, if one or both of the datasets are
purely transformed using filter operations, then instead of rewriting the Except operator
using expensive join operation, we can rewrite it using filter operation by flipping the filter
condition of the right node.
> ![Case-1](https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case1.png)
> ![Case-2](https://github.com/sathiyapk/Blog-Posts/blob/master/images/spark-optimizer/ReplaceExceptWithNotFilter-case2.png)



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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


Mime
View raw message