IBM collaborates with the US government to help fight coronavirus using supercomputers

Collective power of supercomputers should help researchers identify compounds that could be used in the fight against COVID-19

IBM has teamed up with the White House Office of Science and Technology and the US Department of Energy to form a new public/private consortium that will use supercomputers to help fight coronavirus pandemic.

The initiative, named COVID-19 High Performance Computing Consortium, aims to deliver an unprecedented amount of computing power to help researchers better understand COVID-19 and its potential cures.

It will pool supercomputing capacity of various organisations, including IBM, Oak Ridge National Laboratory, Lawrence Livermore National Lab, Sandia National Laboratory, Argonne National Lab, the National Science Foundation, Los Alamos National Laboratory, NASA, and Rensselaer Polytechnic Institute.

In total, 16 supercomputers will be used, offering 775,000 CPU cores, 34,000 GPUs, and more than 330 petaflops of computing power to researchers.

Google, Amazon, and Microsoft are among the members of the consortium.

"These high-performance computing systems allow researchers to run very large numbers of calculations in epidemiology, bioinformatics, and molecular modelling," said Dario Gil, director of IBM Research.

"These experiments would take years to complete if worked by hand, or months if handled on slower, traditional computing platforms."

In addition to offering assistance to American researchers, the consortium will also invite proposals from researchers around the world. The projects that are likely to have an immediate impact will be provided access to supercomputing resources.

According to IBM, the medical researchers at the University of Tennessee and Oak Ridge National Laboratory are already using its IBM Summit supercomputer to identify compounds that can impact the infection process by binding the main spike protein of coronavirus.

So far, the researchers have screened nearly 8,000 compounds in past few days and have identified 77 promising compounds which could potentially impair COVID-19's ability to infect human cells.

At the time of writing, there have been more than 294,110 confirmed cases of COVID-19 worldwide, as per the WHO data, with nearly 12,900 fatalities due to the infection.

Last week, distributed computing project Folding@home (FAH) urged researchers and PC gamers worldwide to donate some of their CPU and GPU computing power to help in the fight against coronavirus.

FAH, which is based in the Stanford University, utilises the idle computing power of hundreds of thousands of PCs owned by volunteers across the world to simulate the molecular dynamics of protein folding and misfolding in various diseases.

Last month, the FAH team announced that it was taking up the fight again coronavirus, with the aim to help develop a therapeutic antibody, similar to that previously developed for SARS-Cov in 2003.

The FAH team is currently aiming to investigate how specific proteins in COVID-19 operate and how those proteins can be destroyed to prevent the virus from multiplying within the human body.