Quando executo o trabalho MPI em InfiniBand, recebo o seguinte desmaio. Nós usamos o Gerenciador de Torque.

--------------------------------------------------------------------------
WARNING: It appears that your OpenFabrics subsystem is configured to only
allow registering part of your physical memory.  This can cause MPI jobs to
run with erratic performance, hang, and/or crash.

This may be caused by your OpenFabrics vendor limiting the amount of
physical memory that can be registered.  You should investigate the
relevant Linux kernel module parameters that control how much physical
memory can be registered, and increase them to allow registering all
physical memory on your machine.

See this Open MPI FAQ item for more information on these Linux kernel module
parameters:

http://www.open-mpi.org/faq/?category=openfabrics#ib-locked-pages

Local host:              host1

Registerable memory:     65536 MiB

Total memory:            196598 MiB

Your MPI job will continue, but may be behave poorly and/or hang.

--------------------------------------------------------------------------

Eu li o link na mensagem de aviso, e eu fiz até agora é;

Acrescentaroptions mlx4_core log_num_mtt=20 log_mtts_per_seg=4 em/etc/modprobe.d/mlx4_en.conf.Certifique-se de que as seguintes linhas estejam escritas/etc/security/limits.conf* soft memlock unlimited* hard memlock unlimitedAcrescentarsession required pam_limits.so em/etc/pam.d/sshdCertificar-se de queulimit -c unlimited é uncommented on/etc/init.d/pbs_mom

Alguém pode me ajudar a descobrir o que estou perdendo?