[
https://issues.apache.org/jira/browse/MAPREDUCE-326?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12834428#action_12834428
]
Doug Cutting commented on MAPREDUCE-326:
----------------------------------------
Arun> We can always revisit this at a later point.
Apache open-source is volunteer-driven. This project should progress in
directions where folks are interested in contributing. One group should not
block avenues of development because they are not their current priority when
they might be a current priority for others.
Owen> By my quick count, you are adding 10 new API classes [ ... ]
That's adding new APIs, not changing existing APIs. The security changes also
add lots of new APIs. New public APIs and new code can be a burden as they
must be supported, so they must provide some clear benefits.
I think, rather than rejecting ideas up-front, we should see if a patch can be
developed that:
- does not slow the existing API
- permits significantly more efficient mapreduce processing in some cases
- permits new, more powerful APIs more efficiently than can be done by layering
If folks wish to create competing patches that accomplish these goals, that'd
be great. This current rash of -1's does not seem productive. I think it
might be time to let code talk for a bit.
> The lowest level map-reduce APIs should be byte oriented
> --------------------------------------------------------
>
> Key: MAPREDUCE-326
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-326
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Reporter: eric baldeschwieler
> Attachments: MAPREDUCE-326-api.patch, MAPREDUCE-326.pdf
>
>
> As discussed here:
> https://issues.apache.org/jira/browse/HADOOP-1986#action_12551237
> The templates, serializers and other complexities that allow map-reduce to
> use arbitrary types complicate the design and lead to lots of object creates
> and other overhead that a byte oriented design would not suffer. I believe
> the lowest level implementation of hadoop map-reduce should have byte string
> oriented APIs (for keys and values). This API would be more performant,
> simpler and more easily cross language.
> The existing API could be maintained as a thin layer on top of the leaner API.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
