The Internet is an ecosystem of about 50,000 Autonomous Systems (or ASes) that operate independently, having different objectives and operational constraints. What glues the Internet together is the bilateral techno-economic agreements that form the interconnections between these ASes. These interconnections (transit, peering, paid-peering, etc) have evolved over the last 20 years or so, since the commercialization of the Internet in the mid-nineties, in a rather ad-hoc manner, often resulting in bilateral or multilateral disputes about who should peer with whom, whether one of the two parties should pay the other, and about the conditions that an interconnection should satisfy (e.g., balanced transit ratios). These problems result in congested interconnections and, in some cases, unreachability problems that can affect millions of Internet users.
In this talk, I will identify a number of key problems in this research area, summarize the state of the art, and describe some of our research results. In particular, the talk will focus on the following questions: 1. Which are the main sources of complexity in Internet peering decisions? 2. How did the peering ecosystem evolve to its current state? 3. How efficient are the contemporary peering practices in economic terms? 4. What is an appropriate way to model the ecosystem of Internet ASes if the focus is on the structure and dynamics of its AS-level interconnections?
Dr. Constantine Dovrolis is a Professor at the College of Computing of the Georgia Institute of Technology. He received the Computer Engineering degree from the Technical University of Crete in 1995, the M.S. degree from the University of Rochester in 1996, and the Ph.D. degree from the University of Wisconsin-Madison in 2001. His current research focuses on cross-disciplinary applications of network science in biology, climate science and neuroscience. He has also worked on the evolution of the Internet, Internet economics, and on applications of network measurement.
