World Economy

US to Build World’s Fastest Supercomputers

US to Build World’s Fastest SupercomputersUS to Build World’s Fastest Supercomputers

The US Energy Department said that it will grant $425 million to research and build two supercomputers, to compete with that of China, that will be the fastest in the world.

The government’s goal is to build computers capable of accurate extrapolation of climate change, improving energy efficiency and they will even provide security for the country’s nuclear weapons.

Approximately $325 million will fund building the supercomputers with parts from IBM, Nvidia and Mellanox. One named “Summit” will be produced at the Energy Department’s Oak Ridge National Laboratory in Tennessee and the other, “Sierra,” at the Lawrence Livermore National Laboratory in California.

“Today’s science is tomorrow’s technology,” said Jen-Hsun Huang, CEO of Nvidia, in a statement. “Scientists are tackling massive challenges from quantum to global to galactic scales. Their work relies on increasingly more powerful supercomputers. Through the invention of GPU acceleration, we have paved the path to exascale supercomputing, giving scientists the tool for unimaginable discoveries.”

After the planned installations in 2017 and 2018, the supercomputers will be five to seven times faster than the most powerful computers in the US. Sierra will run at 100 petaflops, while Summit will reach 150 petaflops. By comparison, today’s faster supercomputer, Tianhe-2 in China, runs at 55 petraflops.

The Tianhe-2, which translates to “Milky Way-2” in English, is so powerful that it required a team of 1,300 scientists and engineers from China’s National University of Defense Technology to build it. The computer is so powerful, that it’s able to perform 33,860 trillion calculations per second.

Another $100 million was awarded to research extreme scale supercomputing in an Energy Department program called FastForward2.

Sierra will be used primarily to simulate nuclear weapon detonation, while Summit will be made available for civilian research, such as computer models for the energy industry.