[jira] Commented: (MAPREDUCE-1135) Making JobHistory log format easily extensible in the future

["Hong Tang (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAPREDUCE-1135?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12768895#action_12768895 ] 

Hong Tang commented on MAPREDUCE-1135:
--------------------------------------

Suggested improvements are:
- Separating serializer and version information from serializer-specific parameters. For instance, whether job history data is encoded in "binary" or "json" format is a parameter to the avro deserializer, and should not be part of the version string.
- Allow the extension of serializer-specific parameters, and possibly allow getting serializer-specific parameters from somewhere else. E.g. we may save many jobs' job history logs into one TFile, and we can save the common schema into TFile's metadata section.

> Making JobHistory log format easily extensible in the future
> ------------------------------------------------------------
>
>                 Key: MAPREDUCE-1135
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1135
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: jobtracker
>    Affects Versions: 0.22.0
>            Reporter: Hong Tang
>
> This jira is branched from MAPREDUCE-1016.
> It would be nice to make the JobHistory log format easily extensible in the future.

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