Re: Octave: Question on running parallel package on linux HPC cluster

[Imran Azeezullah]

a very slight correction (inline)

On 27/09/15 23:53, Carlo De Falco wrote:
> On 25 Sep 2015, at 21:58, Tung Nguyen <address@hidden> wrote:
>
> > Thanks a lot for your input Carlo and Olaf! qsub will give me the computed nodes with 
> > their associated CPUs name. But it's seemed that they all have the same names as in my 
> > first post. I will have to dig further into this to find out how I can access individual 
> > CPUs and use them together with "parallel" package
>
> Tung,
>
> It really depends on what functions you intend to use from the parallel package.
> If you want to use "pararrayfun" and/or "parcellfun", for example, those only 
> work
> if all processes run on the same multiprocessor machine, so all you need to do 
> is
> make sure that the processes you start via qsub are all run on the same machine.
>
> This is usually done in torque by specifying that you request multiple processes
> but only one node, for example to run 16 processes on one node you would write
> something similar in youre launch script:
>
> #!/bin/bash
> #PBS -S /bin/bash
> #PBS -l nodes=1:ppn=1
I think you mean

#PBS -l nodes=1:ppn=16

otherwise you will end up with simply launching multiple threads on that 
single processor you get allocated by PBS.
> cd ${PBS_O_WORKDIR}
> octave --eval "my_parallel_script"
>
>
> > Thanks again for helping!
> >
> > -- Tung
> HTH,
> c.
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/help-octave