Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce >> mail # user >> Scheduling dependent jobs with hadoop-0.20.0


Copy link to this message
-
Re: Scheduling dependent jobs with hadoop-0.20.0
Hi Thomas,

The job control feature has been ported to the new MapReduce API in
https://issues.apache.org/jira/browse/MAPREDUCE-368. This will be in
0.21 I think.

Cheers,
Tom

On Wed, Aug 12, 2009 at 10:55 AM, Thomas Thevis<[EMAIL PROTECTED]> wrote:
> Hi all,
>
> currently, we try to upgrade from hadoop-0.19 to hadoop-0.20. Therefore, we
> try to get rid of the deprecated JobConf class and switch to
> org.apache.hadoop.mapreduce.Job where possible.
> In several projects we have chained jobs which depend on each other. For
> these, we used the Job and JobControl classes of the
> org.apache.hadoop.mapred.jobcontrol package which rely on the deprecated
>  JobConf class.
> How does chaining of dependent jobs work with hadoop-0.20+? Do we still have
> to use the org.apache.hadoop.mapred.jobcontrol package or is there a better
> way using only the org.apache.hadoop.mapreduce.Job class?
> Is there a plan to move org.apache.hadoop.mapred.jobcontrol to a package
> org.apache.hadoop.mapreduce.jobcontrol in the future?
>
> Regards,
> Thomas
>
> --
> Thomas Thevis
> Software Developer
> ------------------------------------------------------------
> vionto GmbH
> Karl-Marx-Allee 90a, D-10243 Berlin
>
> fon   +49 30 40 20 3 29 - 28
> fax   +49 30 40 20 3 29 - 29
> web   http://www.vionto.com
> ------------------------------------------------------------
> Geschäftsführer: Ralf von Grafenstein, Dr. Martin C. Hirsch
> Sitz der Gesellschaft: Berlin
> Amtsgericht Berlin Charlottenburg, HRB 108054B
> ------------------------------------------------------------
>