[
https://issues.apache.org/jira/browse/MAPREDUCE-1473?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Aaron Kimball updated MAPREDUCE-1473:
-------------------------------------
Attachment: MAPREDUCE-1473.patch
Attaching a patch which provides this functionality. This uses
CombineFileInputFormat to batch up Sqoop's input files into a user-defined
number of splits.
As in importing, the degree of parallelism is controlled with the {{\-m}} /
{{--num-mappers}} parameters.
> Sqoop should allow users to control export parallelism
> ------------------------------------------------------
>
> Key: MAPREDUCE-1473
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1473
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: contrib/sqoop
> Reporter: Aaron Kimball
> Assignee: Aaron Kimball
> Attachments: MAPREDUCE-1473.patch
>
>
> Sqoop uses MapReduce jobs to export files back to a table in the database.
> The degree of parallelism is controlled by the number of splits; i.e., the
> number of input files used. The bottleneck in the system, though, is likely
> to be the database itself.
> Users should have the ability to tune the number of parallel exporters being
> used to a degree appropriate to their database deployment.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
