16.7.2012
Sequencial and parallel Dijkstar Implementaiton using OpenMP and OpenMPI
The programs can be compiled using gcc by running
gcc -fopenmp dijkstra_omp.c Dijkstra_tools.c -o dijkstra_omp
and
dijkstra.c Dijkstra_tools.c -o dijkstra
and
mpicc -fopenmp dijkstra_ompMPI.c Dijkstra_tools.c -o dijkstra_ompMPI
There are two valid program arguments. The first specifies the number of Nodes of the graph that will be generated at runtime to test the Dijkstra algoritm, and the second optional argument is the value passed to the srand() function. Rerunning the application with the same both arguments ensures that the same
Matrix is generated!
./dijkstra 1000 19
or
./dijkstra_omp 1000 19
or
mpirun -n 8 ./dijkstra_ompMPI 1000 19
Author: Pasieka Manuel , [email protected]