[torqueusers] Using TORQUE in a supercomputer with lots of CPU's - one node - gets job-exclusive

Jeffrey Lang jrlang at uwyo.edu
Thu Feb 6 10:00:12 MST 2014


   Make sure that you don't have full node allocation turned on in the 
maui configuraton file.

Format: 	one of the following:*SHARED*, *SINGLEJOB*, *SINGLETASK* , or 
Default: 	*SHARED*
Details: 	specifies how node resources will be shared by various tasks 
(See the 'Node Access Overview 
<http://docs.adaptivecomputing.com/maui/5.3nodeaccess.php> ' for more 


(Maui will allow resources on a node to be used by more than one job 
provided that the job's are all owned by the same user)

On 02/06/2014 09:57 AM, Silas Silva wrote:
> Yes.  This is SGI Altix.
> I see, so I have to compile it with --enable-numa-suport?  TORQUE admin
> guide is explanatory about that...
> Thank you!
> On Thu, Feb 06, 2014 at 09:38:45AM -0700, David Beer wrote:
>> Silas,
>> Does this system use NUMA architecture?
>> On Thu, Feb 6, 2014 at 7:34 AM, Silas Silva <silasdb at gmail.com> wrote:
>>> Hi there!
>>> I installed TORQUE and Maui successfully in a supercomputer, with 136
>>> processors.  So, as you can see, this is not a beowulf cluster.  TORQUE
>>> and Maui runs fine, but after the first job is running, the node (the
>>> computer itself) state turns to be job-exclusive, so I can't run any
>>> more job even though there are dozens of processors free.
>>> How to configure it (server_priv/nodes only?) to recognize other CPU's
>>> as a different partition that jobs can be allocated to?
>>> Is multi-mom configuration the answer?
>>> Thank you!
>>> --
>>> Silas Silva
>>> _______________________________________________
>>> torqueusers mailing list
>>> torqueusers at supercluster.org
>>> http://www.supercluster.org/mailman/listinfo/torqueusers
>> -- 
>> David Beer | Senior Software Engineer
>> Adaptive Computing
>> _______________________________________________
>> torqueusers mailing list
>> torqueusers at supercluster.org
>> http://www.supercluster.org/mailman/listinfo/torqueusers

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20140206/985e5d71/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: jrlang.vcf
Type: text/x-vcard
Size: 309 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/torqueusers/attachments/20140206/985e5d71/attachment.vcf 

More information about the torqueusers mailing list