hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xuefu Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (HIVE-9697) Hive on Spark is not as aggressive as MR on map join [Spark Branch]
Date Fri, 20 Mar 2015 03:31:39 GMT

    [ https://issues.apache.org/jira/browse/HIVE-9697?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14370647#comment-14370647
] 

Xuefu Zhang edited comment on HIVE-9697 at 3/20/15 3:31 AM:
------------------------------------------------------------

Thanks, Rui/Chao. So here is what we recommend/conclude for Spark:
{quote}
Spark prefers rawDataSize for map-join memory estimation. Thus, hive.stats.collect.rawdatasize
should be set "true", which is the default. If this configuration is set to false, then fileSize
will be used instead for estimation, which may not be as accurate. 
{quote}
Agree?


was (Author: xuefuz):
Thanks, Rui/Chao. So here is what we recommend/conclude for Spark:
{quote}
Spark prefers rawDataSize for map-join memory estimation. Thus, hive.stats.collect.rawdatasize
should be set "true", which is the default. If this configuration is set to false, then fileSize
will be used instead for memory estimation, which may not be as accurate. 
{quote}
Agree?

> Hive on Spark is not as aggressive as MR on map join [Spark Branch]
> -------------------------------------------------------------------
>
>                 Key: HIVE-9697
>                 URL: https://issues.apache.org/jira/browse/HIVE-9697
>             Project: Hive
>          Issue Type: Sub-task
>          Components: Spark
>            Reporter: Xin Hao
>
> We have a finding during running some Big-Bench cases:
> when the same small table size threshold is used, Map Join operator will not be generated
in Stage Plans for Hive on Spark, while will be generated for Hive on MR.
> For example, When we run BigBench Q25, the meta info of one input ORC table is as below:
>     totalSize=1748955 (about 1.5M)
>     rawDataSize=123050375 (about 120M)
> If we use the following parameter settings,
>     set hive.auto.convert.join=true;
>     set hive.mapjoin.smalltable.filesize=25000000;
>     set hive.auto.convert.join.noconditionaltask=true;
>     set hive.auto.convert.join.noconditionaltask.size=100000000; (100M)
> Map Join will be enabled for Hive on MR mode, while will not be enabled for Hive on Spark.
> We found that for Hive on MR, the HDFS file size for the table (ContentSummary.getLength(),
should approximate the value of ‘totalSize’) will be used to compare with the threshold
100M (smaller than 100M), while for Hive on Spark 'rawDataSize' will be used to compare with
the threshold 100M (larger than 100M). That's why MapJoin is not enabled for Hive on Spark
for this case. And as a result Hive on Spark will get much lower performance data than Hive
on MR for this case.
> When we set  hive.auto.convert.join.noconditionaltask.size=150000000; (150M), MapJoin
will be enabled for Hive on Spark mode, and Hive on Spark will have similar performance data
with Hive on MR by then.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message