[ 
https://issues.apache.org/jira/browse/MESOS-4616?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15136085#comment-15136085
 ] 

Klaus Ma commented on MESOS-4616:
---------------------------------

This is the same requirement of Dynamic Reservation [1] & Persistent Volume 
[2], would you try those two features?

[1] http://mesos.apache.org/documentation/latest/reservation/
[2] http://mesos.apache.org/documentation/latest/persistent-volume/

> Support specifying a preferred host with a Resource Request
> -----------------------------------------------------------
>
>                 Key: MESOS-4616
>                 URL: https://issues.apache.org/jira/browse/MESOS-4616
>             Project: Mesos
>          Issue Type: Story
>            Reporter: Jagadish
>
> When stateful services like Apache Samza, Kafka must be restarted with using 
> Mesos, the framework must have a way of specifying a preferred host with the 
> request. 
> More background:
> I work on Apache Samza , a distributed stream processing framework. Currently 
> Samza supports only Yarn as a resource manager. (there have been requests to 
> run Samza with mesos). A cluster (200 nodes 'ish) runs many Samza Jobs (about 
> 3500). Each Samza Job has its own framework that requests resources 
> (containers) for the job to run. Each such container uses GBs of local state  
> . When such a container(resource) is started on a different host by the 
> framework, the local state must be re-bootstrapped.  (this results in a long 
> bootstrap time, which is essentially down time).
> The same is true for Apache Kafka, a distributed pub-sub logging system.  
> When a Kafka broker must be restarted by the framework, it should ideally be 
> re-started on the same host. (otherwise, each broker has to re-bootstrap 
> several GBs of logs from its peers before it can start to service a request.)
> I'm sure many stateful services have similar requirements.
> My framework can update/ filter resource offers, but with many frameworks 
> (thousands) on several nodes, I'm concerned about the wait-time for a 
> particular available resource to be offered to a framework that needs it.



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

Reply via email to