GAMING
DARPA looks to four organizations to prototype next generations of extreme-scale supercomputers
Military computer scientists at the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va., are looking to industry teams led by NVIDIA in Santa Clara, Calif.; Intel in Hillsboro, Ore.; the Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory in Cambridge, Mass.; and Sandia National Laboratory in Albuquerque, N.M. to develop technologies for a new class of extreme-scale computers that will be 1,000-times more powerful than today's fastest supercomputers.
DARPA also chose Georgia Institute of Technology in Atlanta to lead an applications, benchmarks and metrics team for evaluating the UHPC extreme-scale -- or exascale -- supercomputing systems under development.
DARPA is sponsoring this work as part of its Ubiquitous High Performance Computing (UHPC) program to create an innovative, revolutionary generation of computing systems that overcomes the limitations of current evolutionary approach.
The goal of DARPA’s UHPC program is to re-invent high-performance computing. It plans to develop radically computer architectures and programming models that are 100 to 1,000 times more energy efficient, with higher performance, and that are easier to program than current systems. Prototype UHPC systems are expected to be complete by 2018.
All U.S. Department of Defense (DOD) sensors, platforms, and missions depend heavily on computer systems, DARPA officials point out. The ability to increase performance, however, is limited by high power consumption and complexity, so DARPA is looking to the UHPC program to develop high-performance computers that require less energy per computation.
Computer graphics expert NVIDIA Corp. is leading one of the UHPC teams, which includes the supercomputer company Cray Inc. in Seattle; Oak Ridge National Laboratory in Oak Ridge, Tenn., and six U.S. universities.
"The DARPA UHPC program is attacking technical issues that are key to the future of high performance computing, from the embedded terascale to the exascale," says Steve Scott, Cray's senior vice president, chief technology officer, and the Cray principal investigator on the team.