Traffic Peering Games in Internet Exchange Points

Join a comprehensive lecture by Professor Koushik Kar exploring the dynamics of traffic peering in Internet Exchange Points (IXPs). Delve into the first systematic theoretical study of selfish peering by Internet Service Providers, examining the complex relationship between traffic pricing and flows at IXPs. Learn about congestion-proportional pricing efficiency, port capacity purchases, and queuing delay effects at switch ports. Discover how game theory models demonstrate that equilibrium efficiency remains within a small constant factor of optimum traffic exchange. Explore the automation of peering decisions through machine learning applications using data from PeeringDB and CAIDA repositories. Gain insights into how game theoretic analysis methods can be applied to selfish flow routing, scheduling, and pricing in power and heat flow networks. Professor Kar, an accomplished researcher from Rensselaer Polytechnic Institute, brings extensive expertise in developing decentralized network algorithms and studying resource pricing in shared network environments.