You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-issues@hadoop.apache.org by "Todd Lipcon (JIRA)" <ji...@apache.org> on 2011/06/09 09:53:58 UTC

[jira] [Updated] (HADOOP-7370) Optimize pread on ChecksumFileSystem

     [ https://issues.apache.org/jira/browse/HADOOP-7370?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Todd Lipcon updated HADOOP-7370:
--------------------------------

    Attachment: checksumfs-pread-0.20.txt

I was hacking on this against 0.20. It seems to have made a slight difference but I didn't do any real benchmarking. This would have to be updated to trunk and then benchmarked before commit.

> Optimize pread on ChecksumFileSystem
> ------------------------------------
>
>                 Key: HADOOP-7370
>                 URL: https://issues.apache.org/jira/browse/HADOOP-7370
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: fs
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Minor
>         Attachments: checksumfs-pread-0.20.txt
>
>
> Currently the implementation of positional read in ChecksumFileSystem is verify inefficient - it actually re-opens the underlying file and checksum file, then seeks and uses normal read. Instead, it can push down positional read directly to the underlying FS and verify checksum.

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