Image: Carlos Jones/ORNL via Wikimedia
The White House announced the creation of a supercomputing consortium on Monday that will pool resources to aid in researching and modeling Covid-19. Fourteen different organizations, both federal and private, will give teams of scientists free access to time on their high-powered computing clusters and support from staff.“The sophisticated computing systems available through this Consortium can process massive numbers of calculations related to bioinformatics, epidemiology, and molecular modeling, helping scientists develop answers to complex scientific questions about COVID-19 in hours or days versus weeks or months,” reads the press release.
Supercomputing can aid in modeling and drug discovery by allowing researchers to run a very large number of calculations in a short period of time, enabling high-throughput experiments and precise simulations. The Department of Energy's Summit supercomputer at Oak Ridge National Laboratory alone has a peak speed of 200 petaflops, and according to Quartz it can do as many calculations in a second as every person on Earth doing one calculation a second for the next 6.3 billion years. Summit joins computers from other DOE labs such as Los Alamos and Lawrence Livermore, as well as NASA, private companies like IBM, HP, and Microsoft, and academic institutions. To apply for time on these supercomputers, teams of scientists must submit a two-page proposal summarizing their goals, resource needs, and qualifications. A staff member from the consortium organizations will review and approve applications, and successful teams will be expected to publish their results in scientific journals.
Each computing cluster can be used for slightly different tasks in the researching and modeling of Covid-19. According to the application website, two Department of Energy computing facilities at Argonne National Laboratory and Oak Ridge National Laboratory can be used for “modeling and simulation coupled with machine and deep learning techniques to study a range of areas, including examining underlying protein structure, classifying the evolution of the virus, understanding mutation, uncovering important differences, and similarities with the 2002-2003 SARS virus, searching for potential vaccine and antiviral, compounds, and simulating the spread of COVID-19 and the effectiveness of countermeasure options.”
NASA will be sharing its High-End Computing Capability Portfolio, which the organization uses for large-scale modeling and simulations, while Rensselaer Polytechnic Institute (RPI) will enlist AiMOS, a supercomputer designed to explore applications of artificial intelligence.“We're facing what some might view as an existential threat,” Shirley Ann Jackson, the president of RPI, said in a phone call. “For us to be able to try to help with that, as well as then link it to the epidemiological data and help with the development of interventional strategies, particularly at the therapeutic level, but even at levels that can inform public policy—we feel it's our duty to be part of that.”Jackson said that RPI’s AiMOS will be particularly effective at looking through large databases and, coupled with data analytics, could point to a pathway for successful therapeutics. She added that the 14 groups collectively went to the government and offered their services and time on their supercomputers.“As part of this new effort to attack the COVID-19 pandemic, Hewlett Packard Enterprise is committing to providing supercomputing software and applications expertise free of charge to help researchers port, run, and optimize essential applications in our nation’s time of need,” a spokesperson for Hewlett Packard said in a statement.Supercomputers have been used in past outbreaks, including the 2014 Ebola epidemic and 2015 Zika epidemic. And researchers at Oak Ridge National Laboratory and the University of Tennessee have already used Summit—built by IBM, the fastest supercomputer in the world—to identify 77 candidate small-molecules treatments for Covid-19. The researchers posted their findings as a preprint on the site ChemRxiv.In a blog post about the discovery, IBM research director Dario Gil wrote, “This is the power of accelerating discovery through computation. Now we must scale, and IBM will work with our consortium partners to evaluate proposals from researchers around the world and provide access to this supercomputing capacity for the projects that can have the most immediate impact.”UPDATE: This article was updated with comment from Shirley Ann Jackson, president of RPI.