C++ CUDA implementation of Brande's algorithm for finding betweenness centrality of the graph. Logic follows the solution proposed in Betweenness Centrality on GPUs and Heterogeneous Architectures where authors investigate a set of novel techniques to make BC computations faster with CUDA on GPUs.
"Centrality metrics, such as betweenness and closeness, quantify how central a node is in a network. They have been used successfully to carry analyses for various purposes such as structural analysis of knowledge networks, power grid contingency analysis, quantifying importance in social networks, analysis of covert networks and decision/action networks, and even for finding the best store locations in cities."