CISE Seminar: Vladimir Dvorkin, University of Michigan

Data center coordination with power grids under engineering and privacy constraints

Recent AI breakthroughs have led to significant investments in data center infrastructure, which puts a strain on power grids. As a fundamentally new type of load, AI presents new challenges (e.g., disruptive electricity consumption patterns) and opportunities (e.g., the ability to instantaneously shift computing tasks among spatially distributed data centers). In this talk, I will provide an overview of our research on data center coordination with power grids, specifically addressing the hard feasibility and privacy constraints in this coordination. I will present a new coordination protocol, termed AgentCONCUR, which relies on contextual regression that associates cost-optimal computing task shifts to public and trusted contextual data (e.g., real-time electricity prices and demand) and features minimum data requirements and robust performance guarantees. A New York power grid-based study demonstrates significant coordination gains and identifies optimal contextual features to enable coordination in practice.

Vladimir Dvorkin is an Assistant Professor in the Electrical Engineering and Computer Science Department at the University of Michigan. He has held positions as a postdoctoral fellow at the Massachusetts Institute of Technology’s Energy Initiative and LIDS from 2021–2023, and as a visiting researcher at Georgia Tech’s School of Industrial and Systems Engineering. He received his Ph.D. in Electrical Engineering from the Technical University of Denmark in 2021. His research focuses on the energy transition towards a renewable-dominant and low-carbon energy supply, viewed through the lenses of optimization and machine learning, energy economics, and algorithmic data privacy. His work has received numerous recognitions, including the Marie Skłodowska-Curie Actions and Iberdrola Group postdoctoral fellowship and the IEEE Transactions on Power Systems Best Paper Award.

Faculty Host: Ayse Coskun

Student Host: Mehdi Kermanshah