Supercomputing simulations promise better c-Si understanding

Chinese researchers used a supercomputer to run high-performance molecular simulation, looking to improve production and use of crystalline silicon (c-Si) for photovoltaics and semiconductors.

The Chinese Academy of Sciences' Institute of Process Engineering (CAS-IPE) researchers used Tianhe-1A, which was named the world's fastest supercomputer in 2010 by the Top 500 organization. Tianhe-1A performed a simulation on 7,168 NVIDIA Tesla GPUs, creating 5x better performance than the previous highest-performing molecular simulation, and on a scope twice as big. The simulation modeled the behavior of 110 billion atoms at 1.87 petaflops of performance.

The simulations came "closer than ever" to modeling real crystalline silicon material behavior and true properties in various conditions, said Dr. Wenlai Huang, research associate at CAS-IPE, who notes that the simulations take less time and money than experimental exercises.

Members of the CAS-IPE research team will present the research at the GPU Tech Conference (GTC) in San Jose, CA, May 14-17, 2012, and at GTC China, December 15-16, 2011 in Beijing.

NVIDIA expertise in programmable GPUs has led to breakthroughs in parallel processing that make supercomputing inexpensive and widely accessible.