hadoop-common-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] (HADOOP-8528) Consider using java snappy instead of native
Date Mon, 25 Jun 2012 23:37:44 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-8528?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13401024#comment-13401024

Todd Lipcon commented on HADOOP-8528:

Rather than "instead of", I vote for "as an additional option". I think with some more smarts
about how we handle our buffers going into and out of JNI, the JNI one should end up faster
in the end, but that's an assertion for which I don't have any real data.
> Consider using java snappy instead of native
> --------------------------------------------
>                 Key: HADOOP-8528
>                 URL: https://issues.apache.org/jira/browse/HADOOP-8528
>             Project: Hadoop Common
>          Issue Type: Wish
>            Reporter: Dave Revell
>            Priority: Minor
> Using native libs for snappy has some disadvantages:
>  - Requires libhadoop and libsnappy to be installed on servers, test machines, and any
server running a minicluster with a SNAPPY CF
>  - Requires configuration of java.library.path
>  - Complicates the POM
> A pure java implementation of snappy is available, ASL2 licensed: http://code.google.com/p/snappy-java/
. Benchmarks versus native snappy are available: https://github.com/ning/jvm-compressor-benchmark/wiki

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message