I _think_ what the user is saying is that their "hello world" program is returning rank=0 for all procs when started with mpirun, but not when started with MPICH's mpiexec.hydra.
The most likely problem is that your "hello" program wasn't built against OMPI - are you trying to run the same binary with both mpirun and mpiexec.hydra? If so, that won't work. On Nov 14, 2019, at 8:58 AM, Jeff Squyres (jsquyres) via users <users@lists.open-mpi.org <mailto:users@lists.open-mpi.org> > wrote: Are you asking a question about MPICH? If so, I think you should probably ask on their mailing lists -- they're an entirely different project from Open MPI. Also, I think you mean "processes", not "threads". On Nov 11, 2019, at 5:01 PM, sdcycling via users <users@lists.open-mpi.org <mailto:users@lists.open-mpi.org> > wrote: mpirun -n 4 myprog is only giving me 1 thread when I would like 4. How do I specify 4 threads? mpiexec.hydra -n 4 correctly gives me 4 threads. I have tried to configure a hostfile with no luck. I am running Ubuntu 18.04. Here is a listing of hello.f90: use mpi implicit none integer, parameter :: i4= selected_real_kind(4) integer (i4) :: mype integer (i4) :: nprocs integer (i4) :: ierr call mpi_init(ierr) call mpi_comm_size(mpi_comm_world,nprocs,ierr) call mpi_comm_rank(mpi_comm_world,mype,ierr) write(6,*) 'Hello World',mype call mpi_finalize(ierr) stop end Here is a mpirun example: mpirun -n 4 a.out Hello World 0 Hello World 0 Hello World 0 Hello World 0 Here is a mpiexec.hydra example: mpiexec.hydra -n 4 a.out Hello World 3 Hello World 1 Hello World 2 Hello World 0 Here is the output of lscpu: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 256 On-line CPU(s) list: 0-255 Thread(s) per core: 2 Core(s) per socket: 64 Socket(s): 2 NUMA node(s): 2 Vendor ID: AuthenticAMD CPU family: 23 Model: 49 Model name: AMD EPYC 7742 64-Core Processor Stepping: 0 CPU MHz: 1499.900 CPU max MHz: 2250.0000 CPU min MHz: 1500.0000 BogoMIPS: 4499.95 Virtualization: AMD-V L1d cache: 32K L1i cache: 32K L2 cache: 512K L3 cache: 16384K NUMA node0 CPU(s): 0-63,128-191 NUMA node1 CPU(s): 64-127,192-255 -- Jeff Squyres jsquy...@cisco.com <mailto:jsquy...@cisco.com>
