Haifeng Yu

Haifeng Yu serves as Dean's Chair Associate Professor in the Department of Computer Science at the National University of Singapore's School of Computing. He actively contributes to academic governance as a Member of the Faculty Teaching Excellence Committee (FTEC) and teaches graduate courses including CS4231 Parallel and Distributed Algorithms and CS5223 Distributed Systems.

His educational background includes:

• Ph.D. in Computer Science, Duke University, USA (2002)
• M.S. in Computer Science, Duke University, USA (1999)
• B.E. in Computer Science, Shanghai Jiao Tong University, P.R. China (1997)

Professor Yu's research centers on Distributed Systems Security—particularly blockchain vulnerabilities like sybil attacks—and Distributed Algorithms for dynamic networks. His work bridges theoretical foundations with practical applications in vehicle-to-vehicle communication and disaster recovery systems, where mobile devices form ad-hoc networks when infrastructure fails. Current projects include BCube and Flint for overcoming blockchain's 50% barrier and Massively Parallel Aggregation in dynamic networks.

Analysis of his 2018-2022 publications reveals a dominant focus on blockchain scalability challenges and dynamic network theory. Key trends include developing Byzantine fault tolerance beyond malicious majority thresholds, establishing fundamental lower bounds for network diameter uncertainty, and optimizing sublinear algorithms for T-interval dynamic networks. His work consistently targets top-tier venues like IEEE Security & Privacy (Oakland), JACM, and PODC.

His research excellence is recognized through multiple prestigious awards:

• Best Paper at ACM SPAA (2020)
• Best Paper at ACM SIGCOMM (2010)
• Best Paper at ACM/IEEE IPSN (2009)
• Best Paper at USENIX NSDI (2006)

Professor Yu mentors graduate students including Yuda Zhao and Irvan Jahja, with whom he co-authored seminal works on network diameter costs and dynamic network lower bounds. His research is supported by competitive grants enabling participation in premier conferences where he serves on program committees for PODC, DISC, SIGCOMM, CCS, and Oakland. He leads the BCube/Flint research group investigating blockchain security and dynamic network algorithms, with findings having direct implications for vehicle communication systems and post-disaster recovery networks.