Efficient Graph Comparison and Visualization Using GPU

This paper presents application of several graph algorithms for comparison and visualization of real-world networks. In order to obtain interactive and robust framework for analysis of large graphs we use CUDA implementations of all-shortest-paths (APSP) and breadth-first-search (BFS) algorithms along with CULA matrix decomposition routines. Such an approach allows for efficient computation of graph feature vectors, visualization with graph B-matrices and accelerating dimensionality reduction methods used to embed graphs into low-dimensional metric spaces. Graph analysis algorithms implemented in CUDA were integrated with Graph Investigator Java application via Java Native Interface (JNI) what makes them more convenient to use. We further present two real-world usage scenarios i.e. analysis and visualization of vascular networks in presence of tumor and clusterization based on graph representations of sattelite photos.