Over a year ago, research from Finnish computing center CSC went viral – so to speak – when their simulations showed how virus particles from a cough could spread and linger in the aisles of grocery store. New research led by Michel Boufadel and Fangda Cui of the New Jersey Institute of Technology, however, used the San Diego Supercomputer Center (SDSC) Comet system to demonstrate that grocery stores, if not COVID-free, could at least be less of a viral war zone than previously feared.
The researchers used Comet – which provides 2.76 peak petaflops via 1944 Intel Haswell CPU nodes 72 Nvidia GPU nodes – to simulate virus particles expelled among the grocery islands at a series of time intervals: ten seconds after the release, 100 seconds, 200 seconds and five minutes. In addition, the researchers stratified the study based on the viscosity of these particles.
“We used allowances from the National Science Foundation’s Extreme Science and Engineering Discovery Environment. [XSEDE] for these Comet simulations which allowed us to study the transport of virus-charged particles in an archetypal supermarket 40 meters long by 30 meters wide by 4.5 meters (ceiling height) ”, elaborated Boufadel, professor of civil and environmental engineering at NJIT, in an interview with Kimberly Bruch and Cynthia Dillon of SDSC. “We considered three situations of attachment effectiveness on surfaces: zero percent, 25 percent, and 100 percent attachment. For example, 25 percent attachment efficiency means that out of 100 particles that touch the surface, only 25 percent attached to a surface – so zero percent means no attachment.
The results were surprising: despite the very disparate stickiness of the particles, the 25% and 100% levels had the same fixing “efficiency” when they collided with the shelves, floors and ceilings of the simulated spaces. Overall, these sticky impacts with surfaces reduced the concentration of airborne particles by half, which led Boufadel to cite aisle displays as a useful measure to prevent the spread of COVID in areas. grocery stores (as long as visitors still use hand sanitizer).
Now the team is moving forward to simulate even more minute interactions, like those at the micron level, using additional XSEDE computing resources. “The goal was quantification, and the superior capacity of XSEDE allowed us to have great confidence in the results,” said Boufadel. “XSEDE has opened up a new world to us.”
For more on this research, read the SDSC’s Kimberly Bruch and Cynthia Dillon’s coverage.