Visualizing graph dynamics and similarity for enterprise network security and management

Managing complex enterprise networks requires an understanding at a finer granularity than traditional network monitoring. The ability to correlate and visualize the dynamics and inter-relationships among various network components such as hosts, users, and applications is non-trivial. In this paper, we propose a visualization approach based on the hierarchical structure of similarity/difference visualization in the context of heterogeneous graphs. The concept of hierarchical visualization starts with the evolution of inter-graph states, adapts to the visualization of intra-graph clustering, and concludes with the visualization of similarity between individual nodes. Our visualization tool, ENAVis (Enterprise Network Activities Visualization), quantifies and presents these important changes and dynamics essential to network operators through a visually appealing and highly interactive manner. Through novel graph construction and transformation, such as network connectivity graphs, MDS graphs, bipartite graphs, and similarity graphs, we demonstrate how similarity/dynamics can be effectively visualized to provide insight with regards to network understanding.