[
https://issues.apache.org/jira/browse/SPARK-4352?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14572233#comment-14572233
]
Saisai Shao commented on SPARK-4352:
------------------------------------
Hi [~sandyr], I start to think that shall we need to think this situation where
task number <= executor number * cores, since if dynamic allocation allocation
is enabled, this over-demanded containers will soon be ramped down, also if
container is enough, the container request for adding containers should be
invalid.
So task number >= executor number * cores is the normal situation, and task
number <= executor number * cores should be avoided and will soon be ramped
down, so I think we don't need to consider this situation as special cases,
what do you think? :)
> Incorporate locality preferences in dynamic allocation requests
> ---------------------------------------------------------------
>
> Key: SPARK-4352
> URL: https://issues.apache.org/jira/browse/SPARK-4352
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core, YARN
> Affects Versions: 1.2.0
> Reporter: Sandy Ryza
> Assignee: Saisai Shao
> Priority: Critical
> Attachments: Supportpreferrednodelocationindynamicallocation.pdf
>
>
> Currently, achieving data locality in Spark is difficult unless an
> application takes resources on every node in the cluster.
> preferredNodeLocalityData provides a sort of hacky workaround that has been
> broken since 1.0.
> With dynamic executor allocation, Spark requests executors in response to
> demand from the application. When this occurs, it would be useful to look at
> the pending tasks and communicate their location preferences to the cluster
> resource manager.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
