[torqueusers] C scheduler -> maui help

[etienne gondet]

May be what your are looking for in maui is

NODEACCESSPOLICY SINGLEJOB

JOBNODEMATCHPOLICY     EXACTNODE

just try !

        Etienne Gondet.

Chris Johnson a écrit:

>On Thu, 29 Sep 2005, Chris Samuel wrote:
>
>  
>
>>On Thu, 29 Sep 2005 03:12 am, Chris Johnson wrote:
>>
>>    
>>
>>>     Is there an equivalent to node_pack = False in maui and if so
>>>what?  Thank you.  Much appreciated.
>>>      
>>>
>>This is from an old Maui config file here.
>>
>># Force MAUI to comply with PBS's policy of nodes and jobs
>>JOBNODEMATCHPOLICY     EXACTNODE
>>
>>I think that's what you want..
>>
>>    
>>
>
>     Hm, other Web pages I found seem to indicate this has to do with
>what happens when nodes are specified.  I'm not specifying nodes, just
>letting it pick them.  Node_pack is definitely false.  And
>JOBNODEMATCHPOLICY is definitly EXACTNODE, maui was restarted.  It's
>still ignoring node_pack.  
>
>     Is there a way to force Node_pack=false behavior in maui?  In
>other words if a multiprocessor node is assigned a job, the next job
>should go to a processor on the next node not a processor on the node
>with a job already running.  I believe that's what Node_pack=false
>does.  Or am I wrong?  It seems to be doing this behavior on my
>current cluster.  
>
>-------------------------------------------------------------------------------
>Chris Johnson               |Internet: johnson at nmr.mgh.harvard.edu
>Systems Administrator       |Web:      http://www.nmr.mgh.harvard.edu/~johnson
>NMR Center                  |Voice:    617.726.0949
>Mass. General Hospital      |FAX:      617.726.7422
>149 (2301) 13th Street      |"Quantum mechanics demands that magic exists"
>Charlestown, MA., 02129 USA |        Me
>-------------------------------------------------------------------------------
>
>
>_______________________________________________
>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/20050929/88dd89b9/attachment.html