[
https://issues.apache.org/jira/browse/TAJO-647?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jinho Kim updated TAJO-647:
---------------------------
Attachment: TAJO-647_20140313_17:53:23.patch
> Work unbalance on disk scheduling of DefaultScheduler
> ------------------------------------------------------
>
> Key: TAJO-647
> URL: https://issues.apache.org/jira/browse/TAJO-647
> Project: Tajo
> Issue Type: Bug
> Components: resource manager
> Affects Versions: 0.8-incubating
> Reporter: Jinho Kim
> Assignee: Jinho Kim
> Attachments: TAJO-647.patch, TAJO-647_140312_145207.patch,
> TAJO-647_20140307_14:36:35.patch, TAJO-647_20140313_17:53:23.patch
>
>
> The main problem is that localAllocation does not find next lowest volume.
> and it will assign to remote in rack.
> We should control the remote tasks.
> * Leaf scheduling priorities
> 1. tasks in a volume of host
> 2. unknown disk, non-splitable in local host
> 3. remote tasks in rack. (consider the remaining tasks)
> 4. random remote tasks. (tail concurrency control of scheduled tasks)
--
This message was sent by Atlassian JIRA
(v6.2#6252)
