hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Doug Cutting (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-3315) New binary file format
Date Fri, 09 May 2008 20:47:55 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-3315?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12595716#action_12595716
] 

Doug Cutting commented on HADOOP-3315:
--------------------------------------

 - Magic should be at start of file too.
 - Why use 8-bytes for index length, rowid & offset -- could be VInts of deltas no?
 - Shouldn't metadata be an extensible list of <String,String> pairs?  This could be
encoded as <count>(<string><string>)^count, where <count> is a vint,
and strings are vint-prefixed binary data.  Some metadata keys might be reserved, like "hasKeys",
"hasValues", "keyClass", "valueClass", "entryCount", etc.
 - Need support for total entries.

> New binary file format
> ----------------------
>
>                 Key: HADOOP-3315
>                 URL: https://issues.apache.org/jira/browse/HADOOP-3315
>             Project: Hadoop Core
>          Issue Type: New Feature
>          Components: io
>            Reporter: Owen O'Malley
>         Attachments: Tfile.pdf
>
>
> SequenceFile's block compression format is too complex and requires 4 codecs to compress
or decompress. It would be good to have a file format that only needs 

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


Mime
View raw message