Run-time in seconds: 4 nodes, n qubits, cache size C=2**c

MPICH 3.2.1: 
               n
       c      21      25      31
       7    2.67   37.95 2437.09
       9    0.93   12.24  781.21
      11    0.64    7.41  472.62
      13    0.66    7.43  464.33
      15    0.67    6.37  382.91

Intel MPI Library 2019 Update 3: 

               n
       c      21      25      31
       7    1.11    2.34  105.13
       9    0.82    2.10   91.29
      11    0.83    2.01   90.35
      13    0.81    1.99   86.49
      15    0.81    1.76   82.10

Comparison with Open MPI, n = 31:

       c   MPICH   Intel  OpenMPI
       7 2437.09  105.13   99.48
       9  781.21   91.29   83.97
      11  472.62   90.35   78.89
      13  464.33   86.49   80.86
      15  382.91   82.10   78.15