You are here
Satellite: Joint Analysis of CDNs and Network-Level Interference
Will Scott, Thomas Anderson, Tadayoshi Kohno, and Arvind Krishnamurthy, University of Washington
Awarded Best Student Paper
Satellite is a methodology, tool chain, and data-set for understanding global trends in website deployment and accessibility using only a single or small number of standard measurement nodes. Satellite collects information on DNS resolution and resource availability around the Internet by probing the IPv4 address space. These measurements are valuable in their breadth and sustainability - they do not require the use of a distributed measurement infrastructure, and therefore can be run at low cost and by multiple organizations. We demonstrate a clustering procedure which accurately captures the IP footprints of CDN deployments, and then show how this technique allows for more accurate determination of correct and incorrect IP resolutions. Satellite has multiple applications. It reveals the prevalence of CDNs by showing that 20% of the top 10,000 Alexa domains are hosted on shared infrastructure, and that CloudFlare alone accounts for nearly 10% of these sites. The same data-set detects 4,819 instances of ISP level DNS hijacking in 117 countries.
Open Access Media
USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.