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 Plain View
Hadoop >> mail # dev >> Hadoop Scheduling Algorithm


+
Amr Mohammed 2013-04-21, 18:39
+
Sandy Ryza 2013-04-21, 19:01
+
inelu nagamallikarjuna 2013-04-21, 20:54
+
Chris Smith 2013-04-23, 16:13
Copy link to this message
-
Re: Hadoop Scheduling Algorithm
On 23 April 2013 09:13, Chris Smith <[EMAIL PROTECTED]> wrote:

> And there is another scheduler, Dynamic Priority Scheduling, lurking in the
> backwater of 0.21.0 that allows users to 'bid' for additional time.
> Getting this back into current 1.x may be a great way to understand about
> scheduling:
>
>
> http://svn.apache.org/viewvc/hadoop/common/tags/release-0.21.0/mapreduce/src/contrib/dynamic-scheduler/
>
> Chris
>
>
This is the one that came from HP Labs; it also pre-empts jobs if you
outbid them. It's a lot simpler than the two main schedulers used in
production and should be easier to understand. I don't think it had any
dependencies on 0.21 features; that was just where trunk was when I
committed it
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