Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
MapReduce >> mail # dev >> Split Location.


Copy link to this message
-
Re: Split Location.
On Thu, Mar 17, 2011 at 3:57 AM, Jane Chen <[EMAIL PROTECTED]> wrote:

> Hi,
>
> When a map task fails, and the input split contains multiple locations,
> will the framework choose a different location when re-scheduling the same
> task?
>
> Thanks,
> Jane
>
>
>
> As you say, it might be real. Referring the code
JobInProgress#failMap(TaskInProgress). It adds all the locations with input
split to map cache and expect to run with high priority.
Just the code for your proposal. Look forward to other friends feedbacks and
suggestions!
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB