[
https://issues.apache.org/jira/browse/PIG-4952?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15391286#comment-15391286
]
liyunzhang_intel commented on PIG-4952:
---------------------------------------
After investigating code, the value of parallism can be set by specifying
"parallel":
{code}
A = load './SkewedJoinInput1.txt' as (id,name,n);
B = load './SkewedJoinInput2.txt' as (id,name);
D = join A by (id,name), B by (id,name) parallel 10;
store D into './testJoin.out';
{code}
If user specifies the parallism as 10, then
POGlobalRearrange#requestedParallelism will be 10.
> Parallism is not set as correct value in some physical operator like
> GlobalRearrageParallism is not set as correct value in some physical operator
> like POGlobalRearrange
> -------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: PIG-4952
> URL: https://issues.apache.org/jira/browse/PIG-4952
> Project: Pig
> Issue Type: Sub-task
> Components: spark
> Reporter: liyunzhang_intel
> Fix For: spark-branch
>
>
> Now the parallism value is always 1 in current spark job, this will cause the
> we only have 1 partition in HashPartitioner in shuffle process. see
> [code|https://github.com/apache/pig/blob/spark/src/org/apache/pig/backend/hadoop/executionengine/spark/converter/JoinGroupSparkConverter.java#L90].
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
