Talk by Dr Steve Scott via Teleconference


Nations are eager to develop Exascale systems for reasons of national security and economic competitiveness. Supercomputing systems help accelerate discovery and innovation, benefiting a large number of industries. Supercomputers are also necessary to help solve the world’s most difficult scientific challenges, including finding cures for disease, studying climate change and developing more efficient engines. Exascale systems, the next great challenge in Supercomputing will perform a quintillion floating point calculations per second (that’s a billion billion), making them 1,000 times faster than a one petaflop supercomputer. The world’s fastest computer today is about 16 petaflops. One of the great challenges in developing such systems is in making them energy efficient. Theoretically, an exascale system could be built with x86 processors today, but it would require as much as 2 gigawatts of power — the entire output of the Hoover Dam in the US. Building an exascale system with NVIDIA Kepler K20 processors would consume only 150 megawatts, but this is still an impractical amount. Our goal is to build an exascale system by the end of the decade that consumes just 20 megawatts. This talk will provide an overview of GPU computing and the new NVIDIA Kepler GPUs, discuss the major challenges looming in HPC, and describe NVIDIA’s approach to achieving Exascale computing based on our highly parallel, heterogeneous computing model.

Contact details: 

Subodh Kumar

Department of Computer Science & Engineering