Yes, Intel MPI does support mixed MPI/OpenMP applications. To make sure your hybrid code runs correctly, follow these steps:

1. Use the thread safe version of the Intel MPI Library by using the -mt_mpi compiler driver option
2. Set the I_MPI_PIN_DOMAIN environment variable to select the desired process pinning scheme. The recommended setting is omp:

   $ export I_MPI_PIN_DOMAIN=omp

   This will set the process pinning domain size to be equal to OMP_NUM_THREADS. Therefore, each MPI process can create $OMP_NUM_THREADS number of children threads for running within the corresponding domain. If OMP_NUM_THREADS is not set, each node is treated as a separate domain (which will allow as many threads per MPI processes as there are cores).

   NOTE: In order to pin OpenMP threads inside the domain, use the corresponding OpenMP feature by setting the KMP_AFFINITY environment variable.

3. See the “Interoperability with OpenMP*” section in the Intel® MPI Library for Linux* Reference Manual for more details

Does Intel® MPI Library limit the number of processes run on a single node?

Answer

No, it does not. Intel MPI has no internal restrictions on the number of MPI processes started on a node.

NOTE: Be careful when running a large number of processes on a single machine. At some point, you might start oversubscribing the processor (meaning starting more processes than there are cores). This could lead to performance degradation.