I am quite satisfied with the performance of this code using Julia, as
compared with fortran, matlab and python.
The performance is twice as slow as fortran. The performance gets closer to
fortran when increasing the size of the 2D arrays.
That is in sequential.
The program is very easy to parallelise using OpenMP. I think that it would
be very nice to have such mechanisms with Julia
as laptops/workstations/clusters includes more and more cores.
I want to add also that this code will perform quite well on GPU and MIC
accelerators using OpenMP/OpenACC.
I think that having those Standards (OpenMP/OpenACC) available on Julia
will give it a kick boost !
I implemented this code in order to illustrate scientific computing
language and parallelism
in a course that I will give at Kaust.
Your comments are welcome.
-Alain