[
https://issues.apache.org/jira/browse/MAPREDUCE-961?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Scott Chen updated MAPREDUCE-961:
---------------------------------
Attachment: MAPREDUCE-961-v3.patch
Lots of things changed in this patch.
1. Make MemoryCalculatorPlugin.java collect available memory on the TT.
2. The available memory information is added in
TaskTrackerStatus.ResourceStatus and transmit by heartbeat
3. MemBasedLoadManager launch the task if [Avalable memory - Max memory per
task > Reserved Memory]
where Memory per Task is from existing configuration and reserved memory
needs to be configured.
I have also fixed the problem that Matei pointed out to return
super.canLaunchTask().
@Matei: About the other suggestion about alternatively launching different
types of task, I will open another JIRA and work on that. Because it is not
part of MemBasedLoadManager. Thanks.
> ResourceAwareLoadManager to dynamically decide new tasks based on current
> CPU/memory load on TaskTracker(s)
> -----------------------------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-961
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-961
> Project: Hadoop Map/Reduce
> Issue Type: New Feature
> Components: contrib/fair-share
> Reporter: dhruba borthakur
> Assignee: Scott Chen
> Attachments: HIVE-961.patch, MAPREDUCE-961-v2.patch,
> MAPREDUCE-961-v3.patch
>
>
> Design and develop a ResouceAwareLoadManager for the FairShare scheduler that
> dynamically decides how many maps/reduces to run on a particular machine
> based on the CPU/Memory/diskIO/network usage in that machine. The amount of
> resources currently used on each task tracker is being fed into the
> ResourceAwareLoadManager in real-time via an entity that is external to
> Hadoop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
