hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Min Zhou (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-7339) Introduce a buffered checksum for avoiding frequently calls on Checksum.update()
Date Thu, 09 Jun 2011 01:06:58 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-7339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13046296#comment-13046296

Min Zhou commented on HADOOP-7339:

Hi Nicholas,

Thank you for your feedback. Since right now BufferedChecksum is a non-public inner class,
it should be taken out from DataChecksum as a public class like PureJavaCrc32. 

Could I change the testcase name from org.apache.hadoop.util.TestPureJavaCrc32 to org.apache.hadoop.util.TestChecksum?
Cause it would become a testcase on 4 checksum implementations(or their combinations), jdk
builtin CRC32, PureJavaCrc32, Buffered jdk builtin CRC32, Buffered PureJavaCrc32. 

> Introduce a buffered checksum for avoiding frequently calls on Checksum.update()
> --------------------------------------------------------------------------------
>                 Key: HADOOP-7339
>                 URL: https://issues.apache.org/jira/browse/HADOOP-7339
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: util
>            Reporter: Min Zhou
>             Fix For: 0.23.0
>         Attachments: HADOOP-7339-v1.diff, HADOOP-7339-v2.diff
> We found that PureJavaCRC32/CRC32.update() is the TOP 1 of the methods consuming CPU
in a map side, and in reduce side, it cost a lots of CPU too.
> IFileOutputStream would frequently call Checksum.update() during writing a record. It's
very common a MR key/value less than 512 bytes. Checksum.update() would be called every time
writing a key/value. 
> Test case: terasort 100MB. 
> Checksum.update() calls has be reduced from 4030348 to 28069.  This method is not a hotspot

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message