You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "Bibin A Chundatt (JIRA)" <ji...@apache.org> on 2015/09/01 11:37:45 UTC
[jira] [Updated] (MAPREDUCE-6465) NNBench result wrong when number
of reducers greater than 1
[ https://issues.apache.org/jira/browse/MAPREDUCE-6465?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Bibin A Chundatt updated MAPREDUCE-6465:
----------------------------------------
Attachment: 0001-MAPREDUCE-6440.patch
Thanks [~brahmareddy]
Renaming and attaching patch again.
> NNBench result wrong when number of reducers greater than 1
> -----------------------------------------------------------
>
> Key: MAPREDUCE-6465
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6465
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Bibin A Chundatt
> Assignee: Bibin A Chundatt
> Attachments: 0001-HDFS-8977.patch, 0002-MAPREDUCE-6465.patch
>
>
> Currently NNBench#analyzeResults consider only the part-0000 for analysis
> {code}
> TPS: Create/Write/Close: 0
> Avg exec time (ms): Create/Write/Close: Infinity
> Avg Lat (ms): Create/Write: Infinity
> Avg Lat (ms): Close: NaN
> {code}
> Should consider all part files for output.
> or disable reduces option
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)