Very recently, we witnessed possibly the largest-scale DDoS (Distributed Denial of Service) attack against Spamhaus with attack traffic rates peaked at a whopping 300Gbps. Inspired by this example, I will illustrate a few key challenges facing cyber security. Next I will discuss a truthful mechanism to provide economic incentives to ISPs for participating in distributed defense against large-scale DDoS attacks, and a new Bayesian Network framework for both the attacker and the defender to reason about their strategies in a typical DDoS attack and defense scenario.
Guanhua Yan is a Research Scientist in the Information Sciences Group at Los Alamos National Laboratory. He obtained his Ph.D. in Computer Science from Dartmouth College in 2005. From 2003 to 2005, he was a visiting graduate student in the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign. His Ph.D. dissertation work won the best paper award at PADS'05, a premier conference/workshop on advanced modeling and simulation techniques. His current research interests span various areas of cyber security, particularly cyber genome, anomaly detection in cyberspace, modeling and simulation for cyber security, and cyber infrastructure protection. He has contributed to more than 50 peer-reviewed academic papers and is now the principal investigator of a project that aims to develop phylogenetic methods for malware analysis.