hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vinayakumar B (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-8795) Improve InvalidateBlocks#node2blocks
Date Tue, 21 Jul 2015 08:07:05 GMT

    [ https://issues.apache.org/jira/browse/HDFS-8795?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14634740#comment-14634740
] 

Vinayakumar B commented on HDFS-8795:
-------------------------------------

bq. I think in dump, we don't need to have a sorted result, if agree, I can remove the modification
in dump.
Yes, I also feel there is no need of result to be sorted. 
Since {{node2blocks}} will sort based on datanodes, not the blocks per node, Its fine not
to have sorted result.

> Improve InvalidateBlocks#node2blocks
> ------------------------------------
>
>                 Key: HDFS-8795
>                 URL: https://issues.apache.org/jira/browse/HDFS-8795
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Yi Liu
>            Assignee: Yi Liu
>         Attachments: HDFS-8795.001.patch
>
>
> Currently we use {{TreeMap}} for {{node2blocks}}, actually there are only two place may
need sorted: {{dump}}, {{getDatanodes}}.  But {{getDatanodes}} is called by {{computeInvalidateWork}},
and we do a shuffle there, so the sort is unnecssary.  For {{dump}}, certainly we need to
make few modification.
> So we can use {{HashMap}}.
> From memory and performance view, {{HashMap}} is better than {{TreeMap}}, a simliar optimization
HDFS-7433. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message