hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Konstantin Shvachko (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-415) DFSNodesStatus() should sort data nodes.
Date Fri, 04 Aug 2006 00:23:14 GMT
    [ http://issues.apache.org/jira/browse/HADOOP-415?page=comments#action_12425644 ] 
            
Konstantin Shvachko commented on HADOOP-415:
--------------------------------------------

DatanodeDescriptor is supposed to be used only be the namenode internally.
The class for external use is DatanodeInfo. See HADOOP-321.
That is why DatanodeInfo is public and DatanodeDescriptor is not.
So using DatanodeInfo in dfshealth.jsp was actually right.
I should add some comments in JavaDoc on that.

> DFSNodesStatus() should sort data nodes.
> ----------------------------------------
>
>                 Key: HADOOP-415
>                 URL: http://issues.apache.org/jira/browse/HADOOP-415
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>    Affects Versions: 0.4.0
>            Reporter: Konstantin Shvachko
>             Fix For: 0.5.0
>
>         Attachments: refactor_nodestatus.patch
>
>
> HADOOP-392 introduced a new TreeMap member (datanodeMapByName) in the name node.
> It is used solely for reporting the data nodes in the UI sorted by their names.
> I think it is quite inefficient both time and space-wise to support an excessive
> data structure just for that.
> Suppose the UI will also require sorting by last heartbeat and/or by the available space....
> I think DFSNodesStatus() should just sort the original datanodeMap before returning the
list.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message