You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hama.apache.org by "Thomas Jungblut (Commented) (JIRA)" <ji...@apache.org> on 2011/11/10 00:03:51 UTC

[jira] [Commented] (HAMA-471) Peer names should be sorted by task id

    [ https://issues.apache.org/jira/browse/HAMA-471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13147349#comment-13147349 ] 

Thomas Jungblut commented on HAMA-471:
--------------------------------------

Generally, the task id will be part of the peername znode in zookeeper (byte array stuff).
Then I read things into a treemap (key taskid, value real peername), pop out the values one by one and but it into an array.

I provide a patch tomorrow.
                
> Peer names should be sorted by task id
> --------------------------------------
>
>                 Key: HAMA-471
>                 URL: https://issues.apache.org/jira/browse/HAMA-471
>             Project: Hama
>          Issue Type: Improvement
>            Reporter: Thomas Jungblut
>            Assignee: Thomas Jungblut
>
> Like I already thought in here https://issues.apache.org/jira/browse/HAMA-443?focusedCommentId=13146953&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13146953
> We need to sort the peer names by task id, because this is the only "constant" thing in this system. And it is consistent with the partitioning id, which makes the examples a lot more easier.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira