You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@uima.apache.org by "Jerry Cwiklik (JIRA)" <de...@uima.apache.org> on 2013/12/20 18:48:30 UTC

[jira] [Updated] (UIMA-2667) DUCC Resource Manager defragmentation problem

     [ https://issues.apache.org/jira/browse/UIMA-2667?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jerry Cwiklik updated UIMA-2667:
--------------------------------

    Fix Version/s: 1.0-Ducc

> DUCC Resource Manager defragmentation problem
> ---------------------------------------------
>
>                 Key: UIMA-2667
>                 URL: https://issues.apache.org/jira/browse/UIMA-2667
>             Project: UIMA
>          Issue Type: Bug
>          Components: DUCC
>            Reporter: Jim Challenger
>            Assignee: Jim Challenger
>             Fix For: 1.0-Ducc
>
>
> During defrag a collection of candidate "rich" users and "rich" jobs is made, which forms a collection of candidate shares to evict to make space for NEEDY jobs.  There is a case where a "rich" job may be very rich and yet allocated entirely on machines unsuitable for any NEEDY job.  The bug to fix, is to properly identify this situation and remove the unsuitable shares from consideration.



--
This message was sent by Atlassian JIRA
(v6.1.4#6159)