Presenter: Thom Dunning, U of Washington
Time: 1:00 p.m. - 2:00 p.m.
Place: Watts Family Innovation Center Auditorium (streamed from NIST)
Date: Tuesday, 10/04/2016

Abstract
A new generation of supercomputers—petascale computers—is providing scientists and engineers with the ability to simulate a broad range of natural and engineered systems with unprecedented fidelity. Just as important, in this increasingly data-rich world, these new computers also allow researchers to manage, integrate and analyze unprecedented quantities of data, seeking connections, patterns and knowledge. The impact of this new computing capability will be profound, affecting science, engineering and society.

In 2013, the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign, with funding from the National Science Foundation, deployed a computing system that can sustain one quadrillion calculations per second on a broad range of science and engineering applications as well as manage and analyze petabytes of data. This computer, Blue Waters, has been configured to solve the most compute-, memory- and data-intensive problems in science and engineering. It has tens of thousands of chips (CPUs & GPUs), more than a petabyte and a half of memory, tens of petabytes of disk storage, and hundreds of petabytes of archival storage.

But, computer technology continues to move forward with the U.S. Department of Energy planning to deploy computer systems with peak performances of 100s of PFLOPs next year and 1,000s of PFLOPs in the early 2020s. However, this increase in performance can only be achieved with significant changes in the underlying computing technologies. This presents both and opportunity and a challenge for computational scientists and engineers. The presentation will describe these leading-edge computing systems, illustrate the role that they play or will play in a few areas of research, and describe the challenges facing the development of exascale modeling and simulation.