We present a novel technique, IMaps, for visualizing the Internet topology based on Autonomous System (AS) connections and analyzing traffic metrics and anomalies. The main component of this visualization technique is a geographic-like map that depicts the AS topology. This map represents world countries as contiguous regions of the plane, and ASes and their connections with a node-link diagram. Shape, dimension and boundaries of the countries do not reflect geography, but rather analogue characteristics in the Internet space. Heatmaps are then overlaid to visualize various types of dynamic data, such as aggregate network activity statistics. To analyze AS nodes and detect network anomalies, IP traffic is aggregated into AS nodes and we employ analysis per-AS, so results (and AS heatmap colors) indicate the presence (or not) of anomalies per AS.
