MPI (Message Passing Interface) is the primary parallel run mode used by CFX. MPI is a library that enables flow solver processes to communicate with each other. Proprietary, vendor specific versions of MPI are available on 64-bit Linux and support a wide array of high-speed network devices, as well as Shared Memory communication.

General information on setting up a parallel run and advice on obtaining optimal parallel performance is available in Using the Solver in Parallel in the CFX-Solver Modeling Guide.

Individual machines may need to be configured to run in parallel.