Title: Research Opportunities in Simulation and Artificial Intelligence (AI)
Abstract: A range of domains and applications are poised to significantly gain from recent advances in computer science, particularly in relation to trends in high-end computing and artificial intelligence (AI). In this talk, an overview of the U.S. Department of Energy (DOE)’s Advanced Scientific Computing Research (ASCR) program will be presented, and some areas of opportunity identified by the ASCR research community toward advancing science in the DOE mission will be outlined. With an emphasis on discrete event simulation, opportunities for revisiting key techniques and methods inspired by recent AI approaches and capabilities will be identified, especially in the modeling, usability, and parallel execution aspects. The potential for new research will be highlighted in the confluence of new hardware platforms, software ecosystem, application needs, and algorithmic techniques and methods that are rapidly evolving.
Bio: Kalyan Perumalla is a Program Manager in the ASCR Program of the Office of Science within the U.S. Department of Energy (DOE). Before joining DOE, he spent 17 years in research and managerial roles up to Distinguished Research Staff Member at the Oak Ridge National Laboratory and, prior to that, held research appointments for 8 years at the Georgia Institute of Technology. He was a Fellow of the Institute of Advanced Study at Durham University, UK, served as Joint Full Professor in Industrial and Systems Engineering at the University of Tennessee, and was adjunct faculty at Georgia Tech and the University of Nebraska. He was the recently elected chair of the Association for Computing Machinery (ACM) Special Interest Group in Simulation (SIGSIM) for the terms 2020-2024.
His major areas of technical contributions include scalable parallel/distributed software systems, high-performance computing, GPU-accelerated computing, parallel discrete event simulation, reversible computing, cyber-physical systems, and applications of machine learning. He is the author of “Introduction to Reversible Computing”, a book on the fundamental theory relating energy to computation. Five of his co-authored papers received best-paper awards. He has scaled his algorithms to over 200,000 processor cores and over 8000 GPUs on some of the world’s largest supercomputing systems, significantly pushing the scale and speed of computing in epidemiology, electric grids, cyberinfrastructure, and other domains. Dr. Perumalla earned his Ph.D. in computer science from Georgia Tech in 1999 and was among the first cohort of recipients of the DOE Early Career Award.