tgravescs commented on a change in pull request #26682: [SPARK-29306][CORE]
Stage Level Sched: Executors need to track what ResourceProfile they are
created with
URL: https://github.com/apache/spark/pull/26682#discussion_r371328137
##########
File path: core/src/main/scala/org/apache/spark/resource/ResourceUtils.scala
##########
@@ -232,6 +274,37 @@ private[spark] object ResourceUtils extends Logging {
resourceInfoMap
}
+ /**
+ * This function is similar to getOrDiscoverallResources, except for it uses
the ResourceProfile
+ * information instead of the application level configs.
+ *
+ * It first looks to see if resource were explicitly specified in the
resources file
+ * (this would include specified address assignments and it only specified
in certain
+ * cluster managers) and then it looks at the ResourceProfile to get
Review comment:
Not currently unless those resource managers are lying. Right now for
instance if you request GPUs from yarn and its not configured for it, it throws
an exception. I thought the same was true on k8s and we don't support on
mesos. On the yarn side in the YarnAllocator we also set the resources when we
get the matching requests so yarn would have to be telling us that the
container it gave us matches with those resources.
If it did give us one and we accepted it, then the executor wouldn't start
up if it didn't discovery enough resources that match the resource requirement.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
