[ https://issues.apache.org/jira/browse/HADOOP-2429?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12570862#action_12570862 ]

eric baldeschwieler commented on HADOOP-2429:

The framework should supply a byte ordered comparison by default. This could be highly optimized. It should also support plugable comparators.

I've talked to other users who have done this and seen great improvements in performance.
The lowest level map-reduce APIs should be byte oriented

Key: HADOOP-2429
URL: https://issues.apache.org/jira/browse/HADOOP-2429
Project: Hadoop Core
Issue Type: Improvement
Components: mapred
Reporter: eric baldeschwieler

As discussed here:
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.

Search Discussions

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupcommon-dev @
postedFeb 20, '08 at 10:38p
activeFeb 20, '08 at 10:38p

1 user in discussion

eric baldeschwieler (JIRA): 1 post



site design / logo © 2022 Grokbase