Skitter uses light-weight probing to measure forward IP paths from a source to many (tens of thousands of) destinations, including round trip times and routing changes. By ('traceroute'-style) probing paths to many destination hosts spread throughout the IPv4 address space, we hope to depict a significant portion of Internet connectivity as a directed graph from the skitter host. One of skitter's design goals is to sustain continuous monitoring of many paths in a short period of time, while imposing as little traffic on the network as possible.

Skitter data results in a spanning tree structure originating at the polling host and extending into the infrastructure toward the destination hosts in the polling set. Data is then aggregated into a centralized database for correlation and depiction as a top-down, macroscopic view of the Internet. Juxtaposition of such data sets has been a remarkably unattended area given the promise it offers for insights into the infrastructure as a whole.

The visualization here is an example skitter data run with each link representing a hop between two Internet nodes (IP addresses). We have highlighted links of address space primarily associated with ASNs of ISPs represented at ISMA 98. Note that we color links rather than nodes, and since each link bears two IP addresses, we must select one of those addresses to correspond to the link (as one might expect, it makes a significant difference in the visualization). Each link is directed (all sourced from same CAIDA host, labeled in graph, out toward periphery of image), and so we select the 'to' side of this link to map to a color.

Skitter also tracks approximate round trip times (RTT) and variances. Three-dimensional graphs of RTT variance matrices provide a method to indicate regions of the infrastructure experiencing abnormal delay.

The data derived through skitter will eventually be integrated with a comprehensive database of physical topologies (e.g., prototypes are CAIDA's java-based topology mapping tools for ISP backbones, the Mbone and caching hierarchy topologies). CAIDA is also maintaining a database of latitude and longitude values for identifying physical location of nodes.

Skitter also offers promise in potential correlation to BGP data, to allow engineers to discern who is announcing what to whom over specific paths. Although such information will not answer why this is happening or if such traffic behavior is optimal, it will provide real-world inputs to traffic models/simulations designed to answer such questions. These data can also help pinpoint routing instabilities and other anomalies, and track their secondary, downstream effects, e.g. on round trip times, availability, packet loss across specific paths. A repository of these data/analyses will significantly enhance our predictive capabilities on the Internet.