You can improve performance by using --bind-to socket or --bind-to numa as
this will keep the process inside the same memory region. You can also help
separate the jobs by using the --cpuset to tell each job which cpus it
should use - we'll stay within that envelope.
On Tue, Aug 12, 2014 at 8:33
Am 12.08.2014 um 16:57 schrieb Antonio Rago:
> Brilliant, this works!
> However I’ve to say that it seems that it seems that code becomes slightly
> less performing.
> Is there a way to instruct mpirun on which core to use, and maybe create this
> map automatically with grid engine?
In the open
Brilliant, this works!
However I’ve to say that it seems that it seems that code becomes slightly less
performing.
Is there a way to instruct mpirun on which core to use, and maybe create this
map automatically with grid engine?
Thanks in advance
Antonio
On 12 Aug 2014, at 14:10, Jeff Squyres
The quick and dirty answer is that in the v1.8 series, Open MPI started binding
MPI processes to cores by default.
When you run 2 independent jobs on the same machine in the way in which you
described, the two jobs won't have knowledge of each other, and therefore they
will both starting bingin
