Floyd Warshall Implementation in C++ and CUDA

#Build Instructions

C++ : FloydWarshall.cpp -O3 -o FW.out CUDA: nvcc FloydWarshall.cu -o FW -arch=sm_89 (Ada)

#Limitation

CUDA version does not support input file yet. Generates matrix of random elements for FW algorithm

#Benchmarks (GPU) Benchmarking on my RTX 4060 machine

Vertex Size	Time(s)
4096	2
8192	16
16384	128

Almost perfect cubic scaling