[
https://issues.apache.org/jira/browse/CHUKWA-433?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12796375#action_12796375
]
Ari Rabkin commented on CHUKWA-433:
-----------------------------------
Did the experiment, didn't get interesting results. For my setup, the namenode
was able to keep up with one file per post, and I didn't observe a latency
penalty.
> file-per-post writer
> --------------------
>
> Key: CHUKWA-433
> URL: https://issues.apache.org/jira/browse/CHUKWA-433
> Project: Hadoop Chukwa
> Issue Type: New Feature
> Components: data collection
> Reporter: Ari Rabkin
> Assignee: Ari Rabkin
> Attachments: CHUKWA-433.patch
>
>
> I want to add a writer that outputs a separate HDFS file for each post. The
> use case is performance testing. It's almost certainly useless for any
> production purpose.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
