[
https://issues.apache.org/jira/browse/HADOOP-4439?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12640501#action_12640501
]
Hemanth Yamijala commented on HADOOP-4439:
------------------------------------------
bq. If a job has not specified any memory requirements, this variable would be
set to the job's configuration, maybe via the Task object.
This value is a cluster configuration that should be the same across all task
trackers. Hence, we were trying to see if this can be configured in the job
tracker and passed on to the task trackers. The only way I could find to do it
is to change the {{Task}} object to hold this variable and write it via RPC
when it is transferred to the tasktrackers.
The other option is to simplify, and say that all tasktrackers will be
configured with this value. This has a chance that there could be errors in
configuration if the admin configures different values for different TTs. But,
in an offline discussion with Devaraj, we felt it was simpler to do it this
way, rather than introduce a new field in the {{Task}} object. I am OK with
this approach. Any objections ?
> Cleanup memory related resource management
> ------------------------------------------
>
> Key: HADOOP-4439
> URL: https://issues.apache.org/jira/browse/HADOOP-4439
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.19.0
> Reporter: Hemanth Yamijala
> Assignee: Hemanth Yamijala
> Priority: Blocker
> Fix For: 0.19.0
>
>
> HADOOP-3759 and HADOOP-3581 introduced memory based resource management. This
> JIRA is to cleanup certain aspects of the two issues that came up while doing
> HADOOP-4035, which is filed to support memory based scheduling
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
