hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-1133) Speedup INode.getPathNames
Date Wed, 05 May 2010 22:38:03 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12864561#action_12864561

Todd Lipcon commented on HDFS-1133:

I don't think you want a split that is escapable - you have to use the new split from HADOOP-6623
to do this correctly.

> Speedup INode.getPathNames
> --------------------------
>                 Key: HDFS-1133
>                 URL: https://issues.apache.org/jira/browse/HDFS-1133
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>            Reporter: Dmytro Molkov
>            Assignee: Dmytro Molkov
>         Attachments: HDFS-split.patch
> Currently the method is implemented with the usage of String.split method. But it is
slow because it does regexp parsing of the string.
> If we were to reuse the StringUtil.split method which linearly goes through the string
and splits it on the delimiter characters it will speed up the method by ~20%. Which is great
for NameNode startup time. We can cut about 20% of startup time with this change.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message