hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chris Douglas (JIRA)" <j...@apache.org>
Subject [jira] Commented: (MAPREDUCE-326) The lowest level map-reduce APIs should be byte oriented
Date Tue, 16 Feb 2010 20:45:28 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-326?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12834453#action_12834453

Chris Douglas commented on MAPREDUCE-326:

Doug, the comments on the design document are trying to avoid the other side of this, where
a patch is developed- even committed- without attention to parts of the framework unknown
to the author.

Nobody is imposing a set of priorities. The argument has always been that there is insufficient
evidence to support this direction, but neither is there sufficient evidence against it to
reject it entirely, hence "revisit later," i.e. admitting the possibility that such evidence
could exist. The objections raised are warning that the maintenance burden of the proposed
API may outweigh its benefits. Dismissing those objections and experience as "unproductive"
is ignoring their premise: the APIs are not the cause. You're ignoring all the other suggestions
and alternatives, closing discussion, then asserting that the competition is _between_ new
APIs- the point about which the disagreement turns- and that it's time to write code.

JIRAs related to security aren't adding a parallel mechanism that might, hypothetically, improve
things in _still unnamed_ cases. That analogy is forced.

> 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.

View raw message