Aurora to Bring the Dawn of the Exascale Supercomputer
It’s fitting that the first exascale supercomputer in the United States—with a performance of one exaflop, or a thousand petaflops—will have the name “Aurora,” the mythical Roman goddess of the dawn. After all, it’s the dawn of a new era in high performance computing.
The US Department of Energy (DOE), Argonne National Laboratory, Intel, and supercomputer manufacturer Cray are collaborating to build the new exascale supercomputer, which will serve the research community. Aurora is expected to be delivered to Argonne National Laboratory in 2021, according to the recent DOE announcement.
What are the projected capabilities? Aurora will be able to perform a quintillion (or a billion billion) calculations per second—making it a million times faster than a high-end desktop computer. Along with the ability to handle both traditional high performance computing and artificial intelligence, the machine will provide research teams with tools to address scientific problems at exascale. Researchers will use Aurora to develop alternative energy sources, medical research (such as mapping connections in the human brain that could lead to reversing traumatic brain injury or the cure of Alzheimer’s disease), improving extreme weather forecasting, and other areas.
The Exascale Computing Project website notes that exascale supercomputers can quickly analyze massive volumes of data, and exascale computing has the potential to deliver approximately five times more performance than the nation’s current most powerful supercomputer, the Summit supercomputer at Oak Ridge National Laboratory.
“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,” said U.S. Secretary of Energy Rick Perry in the announcement.
The $500-million Aurora exascale system will be located at the Argonne Leadership Computing Facility, a U.S. Department of Energy Office of Science User Facility, near Chicago in Lemont, Illinois.
According to the Aurora Early Science Program, 15 proposed projects have been initially selected from universities and national labs for work to be done on Aurora at DOE’s Argonne National Laboratory. The projects will support traditional simulation-based computational science, data-centric and data-intensive computing at leadership scale, and machine learning, deep learning, and other AI areas.
If you’re interested, here’s an idea of the types of skills and performance a quantum workforce will require.