[jira] [Commented] (HDFS-7445) Implement packet memory pool in output stream in libhdfs3

["Colin Patrick McCabe (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HDFS-7445?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14238836#comment-14238836 ] 

Colin Patrick McCabe commented on HDFS-7445:
--------------------------------------------

I don't have a strong preference, but I'd like to see us finish stuff like the hdfs.h interface and the addition of unit tests before we start on this optimization.  I feel like this is something we could easily do after merging to trunk.  I guess it's somewhat up to you, though.

> Implement packet memory pool in output stream in libhdfs3
> ---------------------------------------------------------
>
>                 Key: HDFS-7445
>                 URL: https://issues.apache.org/jira/browse/HDFS-7445
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: hdfs-client
>            Reporter: Zhanwei Wang
>            Assignee: Zhanwei Wang
>
> Implement a packet memory pool instead of allocating packet dynamically. A packet memory pool can guard against overcommit and avoid the cost of allocation for output stream in libhdfs3.



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