Final Project for EC521 High-Performance and Parallel Computing, Spring 2015
The goal of this project was to design and develop Dijkstra’s single-source shortest-path algorithm in C, C with OpenMP, and CUDA, with a focus on analyzing and comparing the performance of these three implementations. Performance was compared in terms of graph size, density, and runtime with a focus on analyzing the reasons for differences in runtime.
Source code for the C, C with OpenMP, and CUDA implementations is available on Github.