[
https://issues.apache.org/jira/browse/HADOOP-13200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15980685#comment-15980685
]
Tim Yao commented on HADOOP-13200:
----------------------------------
Thanks [~jojochuang] for your comments.
{quote}
Given that a DataNode can have thousands of concurrent client connections on a
busy cluster, adding this extra overhead is not a good idea. Plus, this array
will not be updated after initialization. I think we can do a better job than
that.
{quote}
I think this is a quite important reason for the modification of this part of
code. I will improve it using another array of names to avoid the extra
overhead.
{quote}
Still in CodecRegistry constructor. I think you want to continue instead of
break if a coder has conflict. Otherwise you would throw an exception instead
of just log an error message.
{quote}
I am sorry I don't quite get the point. I use break when finding whether the
coder name has conflict. If the same name has been detected, an error message
will be printed and loop for the coder name detection will be ended. This coder
factory will then be ignored and the constructor will still continue to
register the remaining coder factories.
And I will take the other advises and apply a new patch accordingly.
> Seeking a better approach allowing to customize and configure erasure coders
> ----------------------------------------------------------------------------
>
> Key: HADOOP-13200
> URL: https://issues.apache.org/jira/browse/HADOOP-13200
> Project: Hadoop Common
> Issue Type: Sub-task
> Reporter: Kai Zheng
> Assignee: Tim Yao
> Priority: Blocker
> Labels: hdfs-ec-3.0-must-do
> Attachments: HADOOP-13200.02.patch, HADOOP-13200.03.patch,
> HADOOP-13200.04.patch, HADOOP-13200.05.patch, HADOOP-13200.06.patch
>
>
> This is a follow-on task for HADOOP-13010 as discussed over there. There may
> be some better approach allowing to customize and configure erasure coders
> than the current having raw coder factory, as [~cmccabe] suggested. Will copy
> the relevant comments here to continue the discussion.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org
