A newly formed COVID-19 High Performance Computing Consortium is working to harness the power of high-performance computing resources to massively increase the speed and capacity of coronavirus research. The consortium, formed last week, comprises several entities including IBM, the White House Office of Science and Technology Policy, and the Department of Energy.
US Chief Technology Officer Michael Kratsios said in a statement Monday that the nation “is coming together to fight COVID-19, and that means unleashing the full capacity of our world-class supercomputers to rapidly advance scientific research for treatments and a vaccine.”
The consortium will be using an “unprecedented amount of computing power” to look at finding new COVID-19 treatments and ultimately vaccines and a cure, Dario Gil, director of IBM Research, said in a statement.
SEE: Coronavirus: Critical IT policies and tools every business needs (TechRepublic Premium)
Researchers will be given remote access to computing power if their projects are approved by the consortium’s leadership board, which includes tech industry leaders and White House and Energy Department officials. The group has begun accepting research proposals through an online portal.
Supercomputers can solve calculations and run experiments that would take traditional computing systems months or years to complete. Traditional computing systems and data centers function and perform calculations independently.
By contrast, high-performance computers can work together and share calculations back and forth to process information more quickly. Such computers are also especially good for conducting research in areas like epidemiology and molecular modeling because the systems mirror the interconnectivity that exists in nature, Gil said.
The consortium will also connect researchers with top computational scientists to ensure the machines are used as efficiently and effectively as possible. The services and computing power will be provided free to researchers.
The consortium is open to additional membership, and besides IBM, the partners that have already offered up their resources include: Amazon Web Services, Google Cloud, and Microsoft from industry; the Massachusetts Institute of Technology and Rensselaer Polytechnic Institute from academia; the National Science Foundation and NASA from the government; and Argonne, Lawrence Livermore, Los Alamos, Oak Ridge, and Sandia National laboratories.
The 16 systems represent more than 330 petaflops, 775,000 central processing unit cores, and 34,000 graphics processing units, “and counting,” which will essentially help scientists deliver results in hours or days, versus weeks or months, according to Gil.
In his blog post regarding the announcement, Gil said the high-performance computing systems will “allow researchers to run very large numbers of calculations in epidemiology, bioinformatics, and molecular modeling.”
Rensselaer Polytechnic Institute President Shirley Ann Jackson told Nextgov Monday that the school is sharing its own AiMOS, which is the most powerful supercomputer at any private university in America—and it’s rated at a sustained rate of 8 PetaFLOPS, or 8 quadrillion calculations per second.
“It is this type of power that is crucial to model a problem as complex as the threat from this coronavirus,” Jackson said.
Even before the consortium’s launch, Jackson said Rensselaer was already being contacted by other entities that were looking to collaborate on COVID-19 data analytics and molecular dynamics, according to Nextgov.
The institute had also already been actively working to attract researchers “who can model all facets of this disease, including spread, drug repurposing, and the development of new vaccines,” Nextgov reported Jackson saying.
And Gil said IBM’s Summit supercomputer has already enabled researchers at the Oak Ridge National Laboratory and the University of Tennessee to screen 8,000 compounds to find those that are most likely to bind to the main “spike” protein of the coronavirus, rendering it unable to infect host cells. Researchers were able to recommend the 77 promising small-molecule drug compounds that can now be experimentally tested, he said.
“This is the power of accelerating discovery through computation,” Gil said.