Edge of Innovation

The world’s fastest supercomputer will research dangerous viruses and better batteries

 

The world’s most powerful supercomputers are already being used for data intensive tasks such as climate modelling, human genome mapping, material science, quantum mechanics and computational chemistry. With the ability to perform billions of calculations per second, they have the power and processing performance to make discoveries that human scientists simply can’t.

According to Nature, a supercomputer in San Jose, California “read 100,000 research papers in 2 hours” and “found completely new biology hidden in the data.” While researchers at the University of Illinois Urbana-Champaign used a supercomputer to model the HIV-1 virus, helping to develop new drugs to combat it.

In the future, they could help scientists and engineers develop better battery technology, fight diseases and improve transportation.

Supercomputer climate modelling
A Category 5 hurricane simulated by the Community Earth System Model (CESM) at a 13 km resolution. Image: ALCF/Alan Scott and Mark Taylor, Sandia National Laboratories.

Intel’s Xeon E5 processors already power the Chinese Tianhe-2 that tops the current Top 500 Supercomputing Sites list. The Tianhe-2 features 3,120,000 cores and has a theoretical peak performance of almost 55 petaflops. It’s capable of performing 33,860 trillion calculations per second.

The new Aurora system will be a Cray-built Shasta supercomputer based on Intel’s High Performance Computing (HPC) scalable system framework. This framework combines Intel Xeon Phi processors, new memory technologies, optical signalling (using Silicon Photonics) and the Lustre parallel file system to achieve a performance leap that will be five to seven times more powerful than Tianhe-2.

How powerful? With a peak performance of up to 450 petaflops, Aurora should be capable of performing 180 quadrillion calculations per second.

Aurora, and a smaller 8.5 petaflop system called Theta, are part of a $200 million project for the US Department of Energy’s (DOE) Argonne Leadership Computing Facility (ALCF). Planned for 2018, Aurora will reportedly be 18 times more powerful than the DOE’s current Mira supercomputer (number five on the Top 500 Supercomputing Sites list), while using only 2.7 times the energy.

Mira supercomputer
The US Department of Energy’s current 10 petaflop Mira supercomputer models everything from jet engines to the human body. Image: ALCF.

“The Aurora system will be one of the most advanced supercomputers ever built, and Cray is honored and proud to be collaborating with two great partners in Intel and Argonne National Lab,” said Peter Ungaro, president and CEO of Cray.

“The combination of Cray’s vast experience in building some of the world’s largest and most productive supercomputers, combined with Intel’s cutting-edge technologies will provide the ALCF with a leadership-class system that will be ready for advancing scientific discovery from day one.”

The DOE plans to use Aurora’s processing power to design new materials that could pave the way for longer-lasting batteries; study organisms to develop improved biofuels; create higher-efficiency and quieter engines; and produce new wind turbine designs, reducing noise and improving energy output.

Using Intel’s HPC systems, Aurora is the next step towards the exascale computing systems of the future – exaflop-class (1,000 petaflop) number crunchers that will be capable of performing a billion billion calculations per second. – Dean Evans

Share This Article

Related Topics

Science

Read This Next