US-built ‘Frontier’ becomes world’s fastest supercomputer


Published: Publication date – 16:24, Mon – 30 May 22

San Francisco: The American supercomputer titled “Frontier” on Monday dethroned the Japanese “Fugaku” (developed by the Riken Institute and Fujitsu), as the fastest machine in the world with 1.1 exaflops of performance on the 59th TOP500 list published by a conference. international IT experts.

The Frontier supercomputer at the US Department of Energy’s Oak Ridge National Laboratory is the first to achieve an unprecedented level of computing performance known as exascale, a threshold of one quintillion calculations per second.

Frontier has a theoretical peak performance of 2 exaflops, or two quintillion calculations per second, making it 10 times more powerful than ORNL’s Summit system.

“Frontier ushers in a new era of exascale computing to solve the world’s greatest scientific challenges,” ORNL Director Thomas Zacharia said in a statement.

“This milestone offers just a glimpse of Frontier’s unrivaled ability as a tool for scientific discovery,” he added.

Frontier builds on ORNL’s vast expertise in accelerated computing and will enable scientists to develop technologies critical to the country’s energy, economic and national security, helping researchers solve problems of national importance that were impossible to resolve just five years ago.

Frontier’s speeds exceeded any other supercomputer in the world, including ORNL’s Summit, which is also housed at ORNL’s Oak Ridge Leadership Computing Facility.

Frontier, an HPE Cray EX supercomputer, also claimed the top spot on the Green500 list, which rates the power consumption and efficiency of commercially available supercomputing systems, with a performance of 62.68 gigaflops per watt.

Frontier’s delivery, installation and testing work began during the Covid-19 pandemic, as shutdowns around the world strained international supply chains.

More than 100 members of a public-private team worked around the clock, from sourcing millions of components, delivering system parts on time, to installing and testing 74 cabinets thoroughly. HPE Cray EX supercomputers, which include more than 9,400 AMD-powered nodes and 90 miles of network cables.

“When researchers gain access to the fully operational Frontier system later this year, it will mark the culmination of work begun more than three years ago involving hundreds of talented people from the Department of Energy and our industry partners. at HPE and AMD,” Jeff said. Nichols, ORNL Associate Laboratory Director for Computing and Computational Sciences.

The exascale performance of the Frontier supercomputer is made possible by some of the world’s most advanced technologies from HPE and AMD.

China has also developed the successors to the Tianhe-2 and Sunway TaihuLight supercomputers.

Source link


Comments are closed.