[
https://issues.apache.org/jira/browse/YARN-1198?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14068594#comment-14068594
]
Chen He commented on YARN-1198:
-------------------------------
{quote}
With preemption, resource beyond guaranteed resource will be likely preempted.
It should be consider as a temporary resource.
{quote}
One thing needs to be clarified about preemption. I think we can resolve
YARN-2008 without introducing preemption. Because if we allow preemption before
define priority, it is wasting time and resource to let thousand of AMs to
compete those "temporary resources" repeatedly.
priority is the most important factor in preemption of scheduling. I think, in
this JIRA, we are talking about how to efficiently and relatively accurate get
headroom in capacity scheduler. Preemption is another story. Here is how
preemption defined in scheduling:
"In computing, preemption is the act of temporarily interrupting a task being
carried out by a computer system, without requiring its cooperation, and with
the intention of resuming the task at a later time. Such a change is known as a
context switch. It is normally carried out by a privileged task or part of the
system known as a preemptive scheduler, which has the power to preempt, or
interrupt, and later resume, other tasks in the system." refer to Preemption
from wikipedia [http://en.wikipedia.org/wiki/Preemption_%28computing%29]
> Capacity Scheduler headroom calculation does not work as expected
> -----------------------------------------------------------------
>
> Key: YARN-1198
> URL: https://issues.apache.org/jira/browse/YARN-1198
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Omkar Vinit Joshi
> Assignee: Omkar Vinit Joshi
> Attachments: YARN-1198.1.patch
>
>
> Today headroom calculation (for the app) takes place only when
> * New node is added/removed from the cluster
> * New container is getting assigned to the application.
> However there are potentially lot of situations which are not considered for
> this calculation
> * If a container finishes then headroom for that application will change and
> should be notified to the AM accordingly.
> * If a single user has submitted multiple applications (app1 and app2) to the
> same queue then
> ** If app1's container finishes then not only app1's but also app2's AM
> should be notified about the change in headroom.
> ** Similarly if a container is assigned to any applications app1/app2 then
> both AM should be notified about their headroom.
> ** To simplify the whole communication process it is ideal to keep headroom
> per User per LeafQueue so that everyone gets the same picture (apps belonging
> to same user and submitted in same queue).
> * If a new user submits an application to the queue then all applications
> submitted by all users in that queue should be notified of the headroom
> change.
> * Also today headroom is an absolute number ( I think it should be normalized
> but then this is going to be not backward compatible..)
> * Also when admin user refreshes queue headroom has to be updated.
> These all are the potential bugs in headroom calculations
--
This message was sent by Atlassian JIRA
(v6.2#6252)
