I have a crisp C / C ++ application. This is basically the main loop for different data sets. We got access to a 100 node cluster with openmp and mpi. I would like to speed up the application, but I am an absolute beginner for mpi and openmp. I'm just wondering what is the easiest way to learn and debug, even if the performance is not the best.
I also wonder what is most suitable for my main loop application.
thank
If your program is just one big loop using OpenMP, it will be as simple as writing:
#pragma omp parallel for
OpenMP , , - kerrighed, , OpenMP node .
MPI . , , .
, " ", , " ", , 100 , node, , 100- , node.
, , 1 "vanilla", " =" . ( Condor, , , sge ..)
OpenMP SMP-, , , MPI. , . MPI OpenMP parallelism node. , OpenMP , pthreads. , , OpenMP, , pthreads.