[
https://issues.apache.org/jira/browse/MAPREDUCE-3816?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13203004#comment-13203004
]
Thomas Graves commented on MAPREDUCE-3816:
------------------------------------------
after thinking about this more. I changed the bar graphs to use the absolute
capacities rather then the relative capacity/max capacity. The reason is that
max capacity is not relative to capacity so having the graph show up properly
when used capacity > capacity when max capacity configured to be > capacity is
not possible in a fashion that shows up both capacity and max capacity
properly.
Another issue with using the relative ones is if max capacity is < capacity:
root queue = 100%
root.test: cap = 20%, max capacity = 100%, absolute cap = 20%, absolute max =
100%
root.test.a1: cap = 80%, max capacity = 20%, absolute cap = 16%, absolute max =
20%
The above should be a valid config but the graph would totally be messed up if
displaying capacity 80% and max capacity 20%.
I also changed used capacity to be what utilization was defined as. I think
this will make more sense to the user and matches what was in 0.20.X/1.0. I
removed utilization and added an absolute used capacity that can be used by the
bar graphs.
While I was in there I also changed it so that instead of opening all the divs
and then closing the leaf queue ones, now everything starts closed and we only
open the parent queues.
> capacity scheduler web ui bar graphs for used capacity wrong
> ------------------------------------------------------------
>
> Key: MAPREDUCE-3816
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3816
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mrv2
> Affects Versions: 0.23.0
> Reporter: Thomas Graves
> Assignee: Thomas Graves
> Priority: Critical
> Attachments: capacityschedulerusedcap.png,
> capschedulerFixedBarGraph.png
>
>
> The capacity scheduler web ui has bar graphs showing the capacity/used
> capacity/max capacity for each queue. The used capacity it is showing is
> actually the % of its parents queue it is using, which doesn't make sense on
> the bar graphs when compared to the capacity and max capacity of that
> particular queue. The bar graphs should be using utilization so that the
> user can see that its using x% or the y% allocated to that queue.
> I will attach some screen shots showing the issue.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
