You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@chukwa.apache.org by "Eric Yang (JIRA)" <ji...@apache.org> on 2009/04/02 00:56:13 UTC
[jira] Updated: (CHUKWA-55) Collect and aggregate slot-hours usage
by user, and by scheduler queues.
[ https://issues.apache.org/jira/browse/CHUKWA-55?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eric Yang updated CHUKWA-55:
----------------------------
Attachment: CHUKWA-55.patch
Slot hours is calculate from sum of the task startup and end time * number of attempts, aggregated for all jobs and group by user. HDFS bytes usage is calculated from dus periodically. By doing time intersection and time grouping, we get a time series data for user usage of the slot hours, and hdfs usage.
> Collect and aggregate slot-hours usage by user, and by scheduler queues.
> ------------------------------------------------------------------------
>
> Key: CHUKWA-55
> URL: https://issues.apache.org/jira/browse/CHUKWA-55
> Project: Hadoop Chukwa
> Issue Type: New Feature
> Components: data collection, Data Processors, input tools, User Interface
> Environment: Redhat 5.1, Java 6
> Reporter: Eric Yang
> Assignee: Eric Yang
> Priority: Blocker
> Attachments: CHUKWA-55.patch
>
>
> The easiest model is:
> Step 1: Collect Disk usage by user
> Step 2: Collect Task Slot usage by user
> Step 3: Aggregating data by scheduler queues
> Step 4: Display data on HICC portal
> Step 5: Profit!
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.