Seyed
Majid
Zahedi,
Department
of
Electrical
and
Computer
Engineering
University
of
Waterloo
Resource sharing is vital to improving efficiency and amortizing cost in high-performance computer systems. Within modern computer systems, users selfishly pursue individual performance without regard for others or the system. To address this challenge and study the strategic behaviour of self-interested users, we turn to algorithmic economics and game theory.
In this talk, we rethink resource management in computer systems, constructing mechanisms that are robust to strategic behaviour. We first discuss some novel methods to manage shared resources in modern data centers. We then talk about game-theoretic learning models and explore their applications in resource management.
Bio: Seyed Majid Zahedi is an Assistant Professor in Electrical and Computer Engineering at the University of Waterloo. His research focuses on the intersection of computer architecture, computer systems, and algorithmic game theory. He is a recipient of multiple paper awards, including the Best Paper Award at HPCA'18 and ASPLOS'16. His paper on computational sprinting has been recently published as a Research Highlight in the Communications of the ACM and his paper on multi-resource allocation was recognized as one of the IEEE Micro Top Picks from Computer Architecture Conferences in 2014.