[jira] [Commented] (SPARK-15176) Job Scheduling Within Application Suffers from Priority Inversion

["Imran Rashid (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-15176?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15311035#comment-15311035 ] 

Imran Rashid commented on SPARK-15176:
--------------------------------------

late to the party, but I support the feature too.  I've seen similar use cases from a number of users, just most haven't gotten sophisticated enough with their deployments to care about this yet (but they probably will soon).

I am still concerned about the what the right api is.  Do we want {{maxShare}} or {{maxRunningTasks}}?  I don't want to add {{maxRunningTasks}} just because its easier to implement, but then we are stuck with the wrong api later on.  (I'm actually not sure I prefer {{maxShare}}, I think I need more time to think about it before I have a strong opinion ... or if somebody else can make a case one way or the other.)

> Job Scheduling Within Application Suffers from Priority Inversion
> -----------------------------------------------------------------
>
>                 Key: SPARK-15176
>                 URL: https://issues.apache.org/jira/browse/SPARK-15176
>             Project: Spark
>          Issue Type: Bug
>          Components: Scheduler
>    Affects Versions: 1.6.1
>            Reporter: Nick White
>
> Say I have two pools, and N cores in my cluster:
> * I submit a job to one, which has M >> N tasks
> * N of the M tasks are scheduled
> * I submit a job to the second pool - but none of its tasks get scheduled until a task from the other pool finishes!
> This can lead to unbounded denial-of-service for the second pool - regardless of `minShare` or `weight` settings. Ideally Spark would support a pre-emption mechanism, or an upper bound on a pool's resource usage.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org