Intel today announced that it and subcontractor Cray will build the first supercomputer with of one exaflop of performance — equivalent to one quintillion floating point computations (“flops”) per second, where a flop equals two 15-digit numbers multiplied together — for the Department of Energy’s Argonne National Laboratory in Chicago. It’s expected to be delivered by 2021.
The Santa Clara company says that the $500 million system, dubbed Aurora, is purpose-built for both traditional high-performance computing and artificial intelligence, and that it will be used to “dramatically” advance scientific research and discovery. It’s the second iteration; Intel previously said it would deploy a 180-petaflop supercomputer at Argonne in 2018, architected on its third-gen Knights Hill Xeon Phi processors, but scrapped the plans after China revealed it intended to build an exascale system by 2020.
At the core of Aurora is a future generation of Intel’s Xeon Scalable processor — Intel Xᵉ — paired with next-gen Optane DC persistent memory. It’ll employ Cray’s Shasta supercomputing system and its Slingshot high-performance interconnect, and fully support Intel’s One API, a suite of developer tools for mapping compute engines to a range of processors, graphics chips, field-programmable gate arrays, and other accelerators.
“There is tremendous scientific benefit to our nation that comes from collaborations like this one with the Department of Energy, Argonne National Laboratory, and industry partners Intel and Cray,” said Argonne National Laboratory director Paul Kearns. “Argonne’s Aurora system is built for next-generation Artificial Intelligence and will accelerate scientific discovery by combining high-performance computing and artificial intelligence to address real-world problems, such as improving extreme weather forecasting, accelerating medical treatments, [charting] the human brain, developing new materials, and further understanding the universe — and that is just the beginning.”
Aurora is an outgrowth of the Energy Department’s Exascale Computing Project (ECP), a grant program within its longrunning PathForward initiative which seeks to accelerate research necessary to develop exascale supercomputers in the U.S. Nearly $258 million in funding was allocated over a three-year contract period starting 2017, and the companies selected to participate — Advanced Micro Devices, Cray, Hewlett Packard Enterprise, IBM, and Nvidia, in addition to Intel — were required to supply supplementary financing amounting to at least 40 percent of their total project cost.
More recently, in April, the Energy Department opened requests for two exoscale systems as part of its CORAL-2 procurement, with a budget ranging from $800 million to $1.2 billion.
The Department of Energy previously awarded $425 million in federal funding to IBM, Nvidia, and other companies to build two supercomputers: one at the Department of Energy’s Oak Ridge and another at Lawrence Livermore National Laboratories. The Oak Ridge system — Summit — delivers between 143 to 200 peak petaflops, according to the TOP500 ranking of supercomputer performance (based on LINPACK score), while Lawrence Livermore’s Sequoia cluster tops out at about 20 petaflops. Both Summit and Sierra were built by IBM and pack IBM Power9 processors and Nvidia Tesla V100 accelerator chips, and consume an enormous amounts of power — up to 13MW, in Summit’s case.
“Achieving exascale is imperative, not only to better the scientific community, but also to better the lives of everyday Americans,” said U.S. Secretary of Energy Rick Perry. “Aurora and the next generation of exascale supercomputers will apply HPC and AI technologies to areas such as cancer research, climate modeling, and veterans’ health treatments. The innovative advancements that will be made with exascale will have an incredibly significant impact on our society.”
Assuming Intel delivers on its promise, Aurora will be the crown jewel in the U.S.’s supercomputer portfolio, but it might not be the world’s most powerful. Three teams in China — in Tianjin (prototype), Jinan, and Beijing — are actively competing to build China’s exascale system in the next seven months, and Japan’s Post-K exascale computer has a target deployment date of 2020.
Currently, America hosts five of the 10 fastest computers in the world, with China’s best — the TaihuLight at the National Supercomputing Center in Wuxi, built on Sunway’s SW26010 processor architecture, and the Tianhe-2A in Guangzhou — ranking third and fourth, respectively, at roughly 125 peak petaflops and 100 peak petaflops. Cray’s Piz Daint sits in fifth ahead of Trinity at Los Alamos National Laboratory, Fujitsu’s AI Bridge Clouding Infrastructure in Japan, and Lenovo’s SuperMUC-NG in Germany.
It’s a fierce arms race betwen China and the U.S. For the first time in TOP500 rankings two years ago, China surpassed the United States in total number of ranked supercomputers, 202 to 143. That trend accelerated in the intervening year; according to the TOP500 fall 2018 report, the number of ranked U.S. supercomputers fell to 108 as China’s total climbed to 229.
China and the United States are followed in the largest number of ranked supercomputers by Japan, which has 31 systems; the U.K., which has 20; France with 18; Germany with 17; and Ireland with 12.