Parallel implementation for Dijkstra algorithm using Java threads
Input data will be taken from "input.txt" and ouput will be put in "output.txt"; it also prints out the execution time in milliseconds.
You can run the jar in the following format:
java -jar ParallelDijkstra.jar [ -seq | (-par <numberOfThreads>) [ <numberOfNodes> <edgeDensity> ]]
Use -seq for sequential or -par for parallel. Input numberOfNodes and edgeDensity if you want to generate a new graph (in which case will be written in "input.txt")
If no arguments are given, the default is a parallel run with 4 threads.
Note: edgeDensity is represented in percentages and needs to be between 0 and 100.
java -jar ParallelDijkstra.jar -par 4 -> will read from "input.txt" for input, execute parallel Dijkstra with 4 threads and write in "output.txt"
java -jar ParallelDijkstra.jar -seq 100 40 -> will generate a graph with 100 nodes and 40% edge density, write it in "input.txt", execute sequential Dijkstra and write the output in "output.txt"
Warning: generating graphs with a large number of nodes may generate huge files (a file containing a graph of 5000 nodes is almost 90MB)