Computing technologies are undergoing a dramatic transition. Because of physical limitations, the frequency of microprocessors, the basis of all computing systems from laptops to supercomputers, has stopped increasing and multiple processor cores are being placed on each microprocessor chip. Dual-core systems were introduced in 2005 and quad-core chips were introduced last year. This will continue, with the number of cores on the chip increasing every 18-24 months. This presents a challenge for computational science and engineering—the only significant performance increases in the future will be through increased use of parallelism. In fact, the use of innovative computing technologies based on many-core chips, e.g., NVIDIA GPUs, are now being explored in computational science and engineering.

At the same time, petascale computers are being planned for several sites across the U.S. The opportunities arising from petascale computing are enormous—predicting the behavior of complex biological systems, understanding the production of heavy elements in supernovae, designing catalysts at the atomic level, predicting changes in the earth’s climate and ecosystems, and designing complex engineered systems from chemical plants to airplanes. However, the challenges are also daunting. Petascale computers are very complex systems, built from multi-core chips with 10,000s of chips and 100,000s of cores, 100s of terabytes of memory, and 10,000s of disk drives. Petascale systems may also be heterogeneous, involving a combination of multi-core and many-core chips. Effective use of petascale computers has significant implications for the design of the next generation of science and engineering applications.

In this presentation, we will provide an overview of the directions in computing technologies as well as describe the petascale computing systems being deployed in the U.S., including Blue Waters—the sustained petascale computing system being funded by the U.S. National Science Foundation at the University of Illinois.