Hello,
I need to decide whether to use v4.1.0 or v4.1.1 for a project and was
wondering if there is any extra information about the latter version. Is
there going to be a rc2 or an ETA for the stable version?
Thanks,
Arturo
Do -cpu-set or -cpu-list work? Or is there a better way to use rankfile?
I have a cluster with 24-cores and 1 GPU per node. I would like to have
one core drive the GPU and the other 23 to be used thread-parallel with
OpenMP. My setup is described in my just-previous email to this list:
CentOS
A couple of BTW's:
The manual mentions "-bind-to-core", but running the code shows that
this is obsolete,
and one must use "-bind-to core".
I can get the bindings by adding --report-bindings:
$ mpirun -np 2 --report-bindings -bind-to core affinity
[vcloud.txcorp.com:3231773] MCW rank 0 bound
OMPI_Affinity_str returning empty strings
I am trying to understand the affinities chosen by OpenMPI
following the documentation of
https://www.open-mpi.org/doc/v4.0/man3/OMPI_Affinity_str.3.php
CentOS-8.2, gcc-8.3, openmpi-4.0.5
$ which mpirun
~/ompi/contrib-gcc830/openmpi-4.0.5-shared/bin/mpir
