[torqueusers] Re: Associating a queue with specific nodes

Anne Hammond hammond at txcorp.com
Fri Jan 5 18:15:03 MST 2007


/var/spool/torque/server_priv/nodes:
node11 np=2 1core opt8
node12 np=2 1core opt8
node13 np=2 1core opt8
node14 np=2 1core opt8
node15 np=2 1core opt8
node16 np=2 1core opt8
node17 np=2 1core opt8
node18 np=2 1core opt8
node19 np=2 1core opt8
node20 np=2 1core opt8
node21 np=2 1core opt8
node22 np=2 1core opt8
...
node24 np=2 1core s3opt24
node25 np=2 1core s3opt24
node26 np=2 1core s3opt24
...

qstat -a
2311.storage2.cl.txc swsides  s3opt24  a005135032    --      5  --    -- 
--  R   --
    node19/1+node19/0+node18/1+node18/0+node17/1+node17/0+node16/1+node16/0
    +node11/1+node11/0



Anne M. Hammond - Systems / Network Administration - Tech-X Corp
                   hammond_at_txcorp.com 720-974-1840

On Sat, 6 Jan 2007, Marc Noguera Julian wrote:

> On Fri, 05 Jan 2007 17:12:02 -0700 (MST), Anne Hammond wrote
>> Thanks marc,
>>
> Hi Anne,
>
>> How do I associate a node in the nodes file
>> (/var/spool/torque/server_priv/nodes) ??
>>
>> pbs is still not allocating the nodes correctly.
>
> the nodes files must be like this:
>
> node1 np=x name_of_queue_1
> node2 np=x name_of_queue_1
> node3 np=x name_of_queue_1
> .
> .
> .
> nodeN np=y name_of_queue_2
> nodeN+1 np=y name_of_queue_2
> nodeN+2 np=y name_of_queue_2
> .
> .
> .
>
> where name_of_queue_1 and _2 correspond to the queues you create via qmgr.
> In this way you add a "queue attribute" to each of the nodes. You can do this
> via qmgr.
>>
>> If, in our submission job, we have:
>>
>> #PBS -q s3opt24
>> #PBS -l nodes=5:ppn=2
>>
>> (Shere nodes23-46 are associated via torque qmgr to queue
>> s3opt24) the job runs on nodes 16-20:
>>
>> 2306.storage2.cl.txc swsides  s3opt24  a005135032    --      5  --
>>  -- --  R
>> --
>>     node20/1+node20/0+node19/1+node19/0+node18/1+node18/0+node17/1+node17/0
>>     +node16/1+node16/0
>>
>> This is not what is supposed to happen.
>>
>> If we use this submission file:
>>
>> #PBS -l nodes=node23:ppn=2+node24:ppn=2+node25:ppn=2+node26:ppn=2+node27:ppn=2
>>
>> the job runs correctly on nodes 23-27.
>>
>> Can anything be done to correct this?
>
> Then, when you submit a job you must also use this "queue attribute" to the
> "PBS -l nodes" instruction.
>
> In your case you are only getting the right use when you ask for specific
> nodes. If you use something like
>
> #PBS -q s3opt24 (assuming s3opt24 is assigned to nodes 23-27 in "nodes" file)
> #PBS -l nodes=5:s3opt24:ppn=2
>
> it should work properly. In this case you are asking for 5 nodes with the
> s3opt24 attribute on them. Note that queue definiton and node specification
> are done in two different ways.
>
> Hope it helps
> Marc
>
> ------------------------------------------------------
> Marc Noguera i Julian, PhD
> Tècnic de suport a la recerca
> Despatx C7-149. Edifici Cn.
> Campus UAB. Bellaterra
> 08193. Barcelona
> email: marc_at_klingon.uab.es
> web: http://klingon.uab.es/marc
> Tlf/Phone: 00 34 935812173
> -------------------------------------------------------
>


More information about the torqueusers mailing list