[
https://issues.apache.org/jira/browse/HADOOP-2233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12546419
]
Doug Cutting commented on HADOOP-2233:
--------------------------------------
> I think this patch needs to be in a src/benchmarks directory
We currently put benchmarks in the src/test tree, which doesn't seem bad.
Benchmarks are a kind of test. And, so long as the class name does not start
with "Test" then they're not picked up by JUnit. I don't see a big need for a
separate tree for benchmarks.
> General example for modeling m/r load in Java
> ---------------------------------------------
>
> Key: HADOOP-2233
> URL: https://issues.apache.org/jira/browse/HADOOP-2233
> Project: Hadoop
> Issue Type: Improvement
> Components: examples
> Reporter: Chris Douglas
> Assignee: Chris Douglas
> Priority: Minor
> Fix For: 0.16.0
>
> Attachments: 2233-0.patch, 2233-1.patch, 2233-2.patch
>
>
> This matches the hadoop.sort.(map|reduce).keep.percent interface in
> HADOOP-2127 and includes an "indirect" sort matching some user apps (when
> locality information is unavailable). It mostly, merely merges parts of the
> RandomWriter and Sort examples to effect sample loads in map/reduce.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
