Climate Supercomputer to Help with COVID-19 Research Efforts
Jacob Bourne posted on May 18, 2020 |
NCAR’s Cheyenne may give unique insights into the pandemic.
Cheyenne supercomputer. (Image courtesy of Carlye Calvin.)
Cheyenne supercomputer. (Image courtesy of Carlye Calvin.)

The National Center for Atmospheric Research (NCAR) announced last month that its Cheyenne supercomputer will be used as part of the COVID-19 High Performance Computing Consortium, an initiative of the U.S. Office of Science and Technology Policy to put computing resources toward addressing the COVID-19 pandemic. The Cheyenne supercomputer, located in Wyoming, is a 5.34-petaflop machine that’s one of the fastest of its kind in the world. The machine is generally used for studying Earth’s systems, particularly the changing climate, and will work alongside the world’s highest-performing computing systems to perform massive calculations related to bioinformatics, epidemiology and molecular modeling to help scientists better understand COVID-19.

“The National Science Foundation is very pleased to be part of the COVID-19 HPC Consortium and provide access to the Cheyenne supercomputer and the NCAR-Wyoming Supercomputing Center,” said Anjuli Bamzai, director of the National Science Foundation (NSF) Division of Atmospheric and Geospace Sciences. “Cheyenne and other NSF-funded high-end computing resources will enable the nation’s research community to pursue advanced modeling using artificial intelligence techniques and other approaches, to gain vital insights into COVID-19 and potential strategies for protecting society.”

Because of Cheyenne’s history of being utilized in Earth science research, the computer is uniquely positioned to help shed light on the intersections between climate change and the COVID-19 pandemic. While there’s no established causal link between the two, scientists are looking at whether changes in regional weather patterns over time could impact how the SARS-CoV-2 virus behaves. Climate change and COVID-19 are both severely urgent issues facing the planet that threaten human life and require solutions on a global level. They’re also issues that are more challenging in terms of mobilizing the general public to address them because the harm they inflict is not instantaneous but rather compounds over time, and both are caused by forces invisible to the human eye.

The modeling done by supercomputers can illuminate the sometimes hard to grasp scope of climate change and COVID-19 through their sheer speed and number processing power. In addition to the NCAR, the consortium consists of several industry, academic and federal agencies, including IBM, Amazon Web Services, Google Cloud, Microsoft, Hewlett Packard Enterprise, MIT, Rensselaer Polytechnic Institute, NSF and NASA.

“Computing and AI have a major role to play in bringing COVID-19 under control,” said Christopher Hill, head of MIT’s research computing infrastructure. “We want to do our part by making MIT’s two most powerful machines, Satori and TX-GAIA, available to researchers who are racing to understand the virus, model the outbreak, and accelerate drug discovery and design. This will be a team effort, and we hope our actions will inspire others to throw their computing power and brain power at the virus.”

The consortium currently has 38 members, 46 projects, and supercomputing systems with a total of 43K GPUs, 437 petaflops and 113K nodes. One of the current projects is examining the impact of atmospheric conditions, particularly temperature and humidity, to determine if the structure of SARS-CoV-2 changes due to atmospheric conditions.

Recommended For You