[
https://issues.apache.org/jira/browse/HADOOP-1986?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tom White updated HADOOP-1986:
------------------------------
Attachment: serializer-v1.patch
Here's a patch (serializer-v1.patch) for discussion. It's still quite
incomplete: there are no tests or javadoc, it produces lots of warnings and I
haven't looked at changing pipes or aggregate to not depend on Writable.
That said, I've managed to run a MapReduce job that doesn't use Writables. (I
created a wrapper round Writable that didn't actually implement Writable, even
though the underlying serialization mechanism used Writable.) The next test is
probably to try using Thrift types.
Thoughts?
> Add support for a general serialization mechanism for Map Reduce
> ----------------------------------------------------------------
>
> Key: HADOOP-1986
> URL: https://issues.apache.org/jira/browse/HADOOP-1986
> Project: Hadoop
> Issue Type: New Feature
> Components: mapred
> Reporter: Tom White
> Assignee: Tom White
> Fix For: 0.16.0
>
> Attachments: SerializableWritable.java, serializer-v1.patch
>
>
> Currently Map Reduce programs have to use WritableComparable-Writable
> key-value pairs. While it's possible to write Writable wrappers for other
> serialization frameworks (such as Thrift), this is not very convenient: it
> would be nicer to be able to use arbitrary types directly, without explicit
> wrapping and unwrapping.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
