sqoop-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Cheolsoo Park (JIRA)" <j...@apache.org>
Subject [jira] [Created] (SQOOP-513) Provide a way to override the default splitter
Date Tue, 03 Jul 2012 22:32:34 GMT
Cheolsoo Park created SQOOP-513:
-----------------------------------

             Summary: Provide a way to override the default splitter
                 Key: SQOOP-513
                 URL: https://issues.apache.org/jira/browse/SQOOP-513
             Project: Sqoop
          Issue Type: Improvement
    Affects Versions: 1.4.1-incubating
            Reporter: Cheolsoo Park


when the number of mappers is greater than 1, Sqoop divides rows using simple queries such
as:

{code}
select x, y from foo where x > 10 and x <= 20.
{code}

The ranges are computed simply by (max - min) / # of mappers. This works fine if values of
the split-by column are distributed evenly; however, it doesn't work well with skewed distribution,
for example.

The proposal is to provide a way so that the user can override the default splitter. For example,
the user should be able to write their own splitter class, pass the class name via a command
option, and use that splitter at runtime instead of the default splitter.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message