Andrew Chien
Abstract Title
Facing the Computing’s Challenges at the End of Technology Scaling – the Power/Energy-Compute Nexxus
While energy is plentiful, the challenges for computing are really about power, because of the localization (space-time) properties of energy. Everywhere you look, the limits of energy on computation have come to the fore. In wearables, in cars, in space, in the cloud, and of course in growing AI datacenter infrastructure. These limits are of course the consequences of the end of Dennard scaling and Moore’s Law. These limits are reframing computer science as an engineering discipline – with limits, arising from physical constraints, economic constraints, and environmental constraints
Today, the macroscopic challenges for computing are about scaling to gigawatts, cost-effectiveness, and minimizing environmental damage. Because cost-effective, carbon-free power supply is dominated by wind and solar generation, the challenge is balancing volatile generation with compute power use. We will describe three approaches: cooperation, decoupling, and as a last resort, compute load flexibility. These should be topics of research at the interface of systems, economics, and optimization. To distribute the costs, a critical question is who will bear the costs of solving these problems – that enable the next 100-fold scaling of computation.. We will also briefly summarize the microscopic compute challenges which are the focus of research on power density, novel devices, compute embedding, and traditional area of microelectronics, micro-cooling, and energy efficiency.
Biography
Dr. Andrew A. Chien is the William Eckhardt Professor of Computer Science at the University of Chicago and a Senior Scientist at Argonne National Laboratory. He leads the IARPA-funded UpDown Graph Supercomputer Project as well as the Zero-carbon Cloud Project. Dr. Chien served as Vice President of Research at Intel Corporation from 2005-2010, leading long-range research and global external research programs. Working with Microsoft and NSF, Chien was instrumental in creation of the Universal Parallel Computing Research Centers (UPCRC) focused on parallel software, the Open Cirrus Consortium (Cloud computing), and Intel’s Exascale Research program. For over 30 years, Chien has been an active researcher in parallel computing, computer architecture, programming languages, networking, clusters, grids, and cloud computing. Previous positions include the SAIC Chair Professor at UCSD (1998-2005), and Senior Fellow SDSC. From 1990 to 1998, Chien was a Professor of Computer Science at the University of Illinois and the National Center for Supercomputing Applications (NCSA) where he was a research leader for parallel computing software and hardware, and developed the well-known Fast Messages, HPVM, and Windows NT Supercluster system that laid foundations for a generation of commodity-based HPC systems.
Dr. Chien is a Fellow of the AAAS, ACM, and IEEE. He has published over 175 technical papers, and his research has been recognized for excellence by numerous awards. Chien received his Bachelor's in electrical engineering, Master's and Ph.D. in computer science from the Massachusetts Institute of Technology.
Contact
aachien@uchicago.edu