There are two versions available. One is for OpenMP shared memory parallelism, useful for running on a single machine. The other version is for MPI to use on a cluster utilizing multiple nodes. The input file in the MPI folder takes a very long time to run, so I recommend changing it before attempting to run it. I have included example Slurm job submissions for GCC and Intel. The Intel file includes an AVX2 requirement that you may need to remove if you don't compile with that flag or don't have the hardware requirements.

I use the clang compiler for the OpenMP version when on MacOS, and it seems to run considerably faster than the GCC counterpart. On Windows, I use GCC. I compile with O3 optimization and for older versions of GCC, you will have to use the 2011 c++ standard flag, along with fopenmp.

On the cluster, I use the Intel compiler with AVX2 and O3 optimization. For OpenMP, use qopenmp and the 2011 C++ standard. For MPI, be sure to compile with impi compatibility if using Intel.