[torquedev] [torqueusers] Problem with ppn and routing

Michel Béland michel.beland at rqchp.qc.ca
Mon Dec 6 15:02:52 MST 2010


Mgr. Šimon Tóth wrote:

>>> Plus the Torque developers still haven't decided on clear resource
>>> semantics therefore -l nodes=2 -l mem=2G could mean both 2GB spread
>>> across 2 nodes or 2GB on each node (4GB total).
>> It seems to me that Torque views mem as the total memory. That is what I 
>> guess from qstat -f from one of the jobs that is running on our cluster. 
>> Here is an excerpt:
>>
>>      resources_used.mem = 182440144kb
>>      resources_used.vmem = 252290712kb
>>      Resource_List.mem = 250gb
>>      Resource_List.nodect = 16
>>      Resource_List.nodes = 16:ppn=4
>>
>> As our nodes have 8 cores and 16 GB, I do not see how mem could be 
>> interpreted as memory per node with the numbers that I get.
> 
> No, torque doesn't view mem at all. The only thing that is done with
> resources like mem are checks against server and queue minimum and
> maximum limits. If no limits are set, no checking is done and resources
> are ignored.

This does not corresponds to my definition of "at all". What else should 
Torque do with a resource in order that one can consider that it views 
it? Pbs_mom passes to the server how much mem is used locally and the 
server sums up the result. The server understands mem enough so that it 
can route jobs according to the value requested. What else should it do?

-- 
Michel Béland, analyste en calcul scientifique
michel.beland at rqchp.qc.ca
bureau S-250, pavillon Roger-Gaudry (principal), Université de Montréal
téléphone : 514 343-6111 poste 3892     télécopieur : 514 343-2155
RQCHP (Réseau québécois de calcul de haute performance)  www.rqchp.ca
Calcul Canada (computecanada.org)


More information about the torquedev mailing list