[Mauiusers] cleaning up "ncpus" mess

Alessandro Federico alessandro.federico at caspur.it
Tue Apr 11 10:16:18 MDT 2006


hi Garrick,

i have filled up my queues with many jobs requesting different
number of cpus. every job script checks the number of nodes
and cpus allocated by maui. no errors were found!
so i'm very happy of your patch and i thank you again.
if i will discover some problems i will write you.

bye
ale


Garrick Staples wrote:
> On Wed, Apr 05, 2006 at 07:23:34PM +0200, Alessandro Federico alleged:
>> thanks for reply Garrick.
>> i've tried to reproduce your fault but it works.
>> can you be more clear about the conditions when it happens?
> 
> I think if you submit a whole bunch of jobs at once, more than can be
> run immediately, you'll see some of the later jobs run with 1 CPU.
> 
> 
>> thank you in advance
>>
>> ale
>>
>> Garrick Staples wrote:
>>> On Mon, Apr 03, 2006 at 06:38:25PM +0200, Alessandro Federico alleged:
>>>> hi Garrick,
>>>>
>>>> i'm trying your patch on my dual Opteron cluster and
>>>> it seems to work. i'm very happy because i plan to use
>>>> torque/maui on our IBM SMP cluster (8 cpus per node)
>>>> where i would like to request cpus (not nodes=X:ppn=Y)
>>>> and let the scheduler choose the nodes.
>>>> on the opteron cluster i discoverd that, if i submit
>>>> a job with -l ncpus=4, maui creates only one task
>>>> (with PROCS=4) and the job will never run because it
>>>> cannot find a node (with 4 cpus) to satisfy the task.
>>>> it seems a strange behaviour!
>>>>
>>>> anyway, thank you very much. i will give you feed back
>>>> as soon as i will try your patch on the IBM.
>>> Yah, about that patch... it didn't work so well.  I found out later that
>>> in some cases Maui allocated 1 CPU to jobs.  I think it happened when
>>> resources weren't available when the job was first submitted; subsequent
>>> attempts to run the job only got 1 CPU.
>>>
>>>
>>>> PS: why did nobody reply you?
>>> *shrug*  because I hardly ever get replies when I ask people to test
>>> stuff.
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> mauiusers mailing list
>>> mauiusers at supercluster.org
>>> http://www.supercluster.org/mailman/listinfo/mauiusers
>> -- 
>> ***************************************************
>>      Alessandro Federico
>>      CASPUR  -  http://www.caspur.it/
>>
>>      e-mail:    alessandro.federico at caspur.it
>>      phone:     +39 06 44486708
>>      fax:       +39 06 4957083
>>
>> ---------------------------------------------------
>>  Military intelligence is a contradiction in terms.
>>                                     (Groucho Marx)
>> ---------------------------------------------------
> 

-- 
***************************************************
     Alessandro Federico
     CASPUR  -  http://www.caspur.it/

     e-mail:    alessandro.federico at caspur.it
     phone:     +39 06 44486708
     fax:       +39 06 4957083

---------------------------------------------------
 Military intelligence is a contradiction in terms.
                                    (Groucho Marx)
---------------------------------------------------


More information about the mauiusers mailing list