Computation of faculty of n

    fac(n) = n * (n-1) * (n-2) * ... * 6 * 5 * 4 * 3 * 2 * 1

    results in

    a(size) a(size-1) ..... a(3)  a(2)  a(1)   number to base 65536

    where every a(i) is between 0 and 65535.

    Output of program is sum of a(1:size)

Some results:

    n        size         control sum
   ===================================
   100         33         816765
  1000        534       15597330
  5000       3390      100792830
 16000      12524      379185510

fakul.f:

    Fortran 90 Version of the algorithm

    5000 (8 processors) : ? s

fakul1.f:

    Realization with forall, the CSHIFT operation remains

    5000 (8 processors) : 12 s

fakul2.f:

    CSHIFT is replaced with node array movement

    5000 (8 processors) : 19 s

