In principal if the queuing system is setup correctly then it should
automatically set OMP_NUM_THREADS to the correct value and any openmp code
should obey that. This should also work with MPI in that if the user asked
for 8 tasks and 1 task per core then they would be running with 8 MPI
threads and OMP_NUM_THREADS would get set to 1. Most queuing systems know
the difference between sockets, cpus and cores these days (although a lot
still get confused over hyperthreads) and so it should in principle just
work.
Of course this is not true on desktop machines or systems without queues
where the user will need to set OMP_NUM_THREADS themselves.
All the best
Ross
On 4/28/13 4:10 AM, "David A Case" <case.biomaps.rutgers.edu> wrote:
>On Sat, Apr 27, 2013, M. L. Dodson wrote:
>>
>>
>> I don't know how widespread openmp support is among operating systems
>>likely
>> to be used for AmberTools installs, but if it is near universal, my
>>opinion
>> is that -openmp installs should be the default, not serial installs.
>>This is
>> because almost all modern CPUs are multicore, hence fit well with
>>openmp.
>
>Our main concern in making this the default is/was that users might be
>using
>multiple cores when they weren't expecting to, potentially violating local
>queue requirements, etc. But I'm cc-ing this to amber-developers, to get
>other opinions. You are correct that certain types of workstation jobs
>would
>indeed be faster if users routinely made us of the openmp capabilties.
>
>...thanks...dac
>
>
>_______________________________________________
>AMBER-Developers mailing list
>AMBER-Developers.ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber-developers
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Sun Apr 28 2013 - 13:30:02 PDT
