Am 22.08.2012 um 21:57 schrieb Joseph Farran:

> The linky Rayson shows below works great.
> 
> I am converting from Torque/Maui to Grid Engine and the one thing that I 
> missed the most in Torque was the ability to easily request "whole nodes".
> 
> I should add that the link below works *BUT* you will not able to use it with 
> Subordinate queues.
> 
> What happens is that once a subordinate queue grabs a node in exclusive node, 
> it wont let it go and so jobs cannot be suspended so that the higher priority 
> queue can run.

Yes, as the suspension is the result of the superordinated queue starting a job 
there. There is no look-ahead in SGE. What about using all cores by intention 
by specifying an integer value in the PE of the subordniated queue? You are 
limited to multiple of a certain value of course.

-- Reuti


> If anyone knows how to grab exclusive nodes with subordinate queues that can 
> be suspend please speak up :-)
> Joseph
> 
> 
> On 08/22/2012 12:42 PM, Rayson Ho wrote:
>> There's Exclusive Scheduling for allocating the whole node:
>> 
>> http://docs.oracle.com/cd/E24901_01/doc.62/e21978/management.htm#sthref431
>> 
>> Rayson
>> 
>> 
>> 
>> On Wed, Aug 22, 2012 at 3:39 PM, Henrichs, Juryk
>> <[email protected]>  wrote:
>>> Hi,
>>> 
>>> we have a heterogeneous cluster consisting of nodes with 32 and 48
>>> cpu's. Some of our jobs are best run either on one complete node or
>>> distributed over a number of nodes. In either case it is desired that
>>> the job runs exclusively on each node and uses all available slots.
>>> 
>>> The reason for this is that we use mpirun with --mca mpi_paffinity_alone
>>> 1 as this gives the best performance.
>>> 
>>> The problem is, that up to now we have to take care on which type of
>>> nodes a job is running and that jobs are not distributed over mixed nodes.
>>> 
>>> Is there a way to submit jobs with a somehow flexible number of slots
>>> such that only complete nodes are allocated?
>>> 
>>> I am aware of the $pe_slots allocation rule, but as I understand it, a
>>> job which requests more slots than available on one single node will
>>> never get started.
>>> 
>>> Juryk
>>> 
>>> 
>>> This e-mail and any attachment thereto may contain confidential information 
>>> and/or information protected by intellectual property rights for the 
>>> exclusive attention of the intended addressees named above. Any access of 
>>> third parties to this e-mail is unauthorised. Any use of this e-mail by 
>>> unintended recipients such as total or partial copying, distribution, 
>>> disclosure etc. is prohibited and may be unlawful. When addressed to our 
>>> clients the content of this e-mail is subject to the General Terms and 
>>> Conditions of GL's Group of Companies applicable at the date of this e-mail.
>>> If you have received this e-mail in error, please notify the sender either 
>>> by telephone or by e-mail and delete the material from any computer.
>>> GL's Group of Companies does not warrant and/or guarantee that this message 
>>> at the moment of receipt is authentic, correct and its communication free 
>>> of errors, interruption etc.
>>> FutureShip GmbH, HRB 106781 AG HH, VAT Reg. No. DE263937825
>>> Geschäftsführer (CEO): Volker Höppner, Henning Kinkhorst, Stefan Deucker
>>> 
>>> 
>>> _______________________________________________
>>> users mailing list
>>> [email protected]
>>> https://gridengine.org/mailman/listinfo/users
>> _______________________________________________
>> users mailing list
>> [email protected]
>> https://gridengine.org/mailman/listinfo/users
>> 
> 
> _______________________________________________
> users mailing list
> [email protected]
> https://gridengine.org/mailman/listinfo/users
> 


_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to