Peak Performance

Designed with general purpose users in mind, the Fugaku supercomputer has the performance and energy efficiency of a GPU-accelerated machine but remains easy to program.

Within weeks, the first results were announced. In July 2020, researchers from RIKEN and Kyoto University said Fugaku had helped them narrow down a field of 2,128 drugs to dozens of potential therapeutics that might bind to and inactivate proteins that play a role in how the virus particles spread. Fugaku took ten days running molecular simulations to find the candidates, some of which had not yet been considered as possible agents.

Apart from screening drug candidates, Fugaku has also helped scientists understand how the coronavirus can spread while airborne. A research team from RIKEN and Kobe University was preparing to use Fugaku to simulate vehicle fuel-injection systems when one member realized that it was basically the same mechanism as coughing.

Working with biologists and other experts, the team ended up running what it described as the world’s largest virus-droplet simulation ever conducted. Instead of atomized fuel, Fugaku was put to work modeling how the coronavirus can spread in trains, workspaces and schools.

They found that opening train windows, for instance, can triple ventilation and lower ambient particles. Another study looked at how particles emitted by infected people would behave in a variety of environmental conditions.

They found that if air humidity is under 30 percent, there are more than twice the number of aerosolized particles compared to levels over 60 percent. Other findings revealed that the biggest risk to diners is people sitting next to, and not across from them, and that face masks are more effective than clear face shields.

“They got probably the best simulation in the world, because they were developing this extremely complicated but capable code that would not work too well on conventional machines because they’re too slow,” said Matsuoka, adding that some simulations that took a few days on Fugaku would have taken over a year on the K computer.

“We had a very focused, very unified response to the pandemic. Everybody contributed according to their abilities and we moved quickly in the same direction.”

Tim Hornyak is a Canadian writer based in Tokyo, Japan, who has worked in journalism for more than 20 years. He has written extensively about travel, food, technology, science, culture and business in Japan, as well as Japanese inventors, roboticists and Nobel Prize-winning scientists. He is the author of Loving the Machine: The Art and Science of Japanese Robots and has contributed to several Lonely Planet travel guidebooks. He has lived in Tokyo for more than 15 years.

Related Stories from Asian Scientist