config-one-proc-per-node

collectives : allgather allgatherv allreduce alltoall alltoallv barrier bcast gather reduce reduce_scatter_block reduce_scatter scatter
omb_collective_directory : /path/to/install/libexec/osu-micro-benchmarks/mpi/collective/
imb_binary : /path/to/install/IMB-MPI1
number_of_ranks : 2 3 4 5 6 7 8
max_num_node : 8
number_of_cores_per_node : 1
number_of_runs_per_test : 3
mpirun_options : --map-by ppr:1:node