RICHLAND, Wash. – Researchers at the U.S. Department of Energy’s Pacific Northwest National Laboratory have been allocated more than 3 million node hours on the nation’s most powerful computers to investigate questions about pathogens, climate and energy-efficient microelectronics.
The access to the nation’s supercomputer given to Margaret Cheung, Daniel Mejia Rodriguez and Poh Lun Ma is a coveted prize among scientists, and each node hour represents a multimillion-dollar investment in computing time given to PNNL scientists to investigate important scientific questions.
The award is one of 44 projects awarded by DOE’s Advanced Scientific Computing Research Program through DOE’s 2024-2025 ASCR Leadership Computing Challenge, which awards computing hours at the Argonne Leadership Computing Facility, Oak Ridge Leadership Computing Facility and the National Energy Research Scientific Computing Center.
Pathogen-host conflict
Margaret Chung uses her expertise in predictive phenomics to explore pathogen-host interactions using 1 million node hours. (Photo by Andrea Starr | Pacific Northwest National Laboratory)
Margaret Chung is using 1 million node hours of computational time to study how viruses infect algae. This process offers clues for preparing for possible future pandemics. The virus-algae system is a window into the broader tug-of-war that is constantly happening between pathogens and hosts. One only has to hear about the latest COVID strains to see that viruses are constantly mutating, going after ever-changing targets. Chung will leverage the resources of DOE’s most advanced computers, including OLCF’s Frontier, the world’s fastest computer, as well as resources at ALCF and NERSC, to unlock the secrets of the pathogen-host battle.
Cheung’s research is part of a program at PNNL called predictive phenomics, a major effort in which scientists make fundamental changes to molecular interactions and monitor how those changes change organisms. For decades, scientists have known that an organism’s genes, or DNA, are very important. They’re finding that there are many other, more subtle biological processes that play a major role in biology. Cheung’s study of such events in virus-algae interactions could be a roadmap for understanding the same processes in humans. The project is supported by PNNL and the DOE’s Biopreparedness Research Virtual Environment (BRaVE) initiative.
“We want to know exactly what’s happening between the pathogen and the host — not just what’s happening, but when, where and how it’s happening,” Chen says. “This involves a lot of fundamental physics and chemistry, even for algae or simple viruses.”
Aerosols and Climate
Poh Lun Ma studies the role of aerosols in climate. (Photo by Andrea Starr | Pacific Northwest National Laboratory)
Geoscientist Poh Loong Ma is focused on better representing atmospheric aerosols in climate models. Aerosols are small particles that naturally occur from wildfires, ocean spray, and pollutants, and are one of the biggest sources of uncertainty in climate simulations. Scientists think that aerosols could cool or warm the Earth, depending on the situation. Ma leads a study called EAGLES, which continues to resolve uncertainties and discover important details about how clouds interact with Earth’s atmosphere.
This is the fourth time Ma has received the award; the previous three awards were also awarded for his work on EAGLES, which this year was awarded 1.6 million node-hours of Frontier computing time.
“We’re proud of our work, but we’re not satisfied yet,” Ma said. “We’re helping to reveal important details that few others on the planet know — details that are key to making informed decisions about the climate-related challenges facing countries and societies around the world.”
Spintronics and Electronic Efficiency
Computational scientist Daniel Mejia Rodriguez works to make digital electronics more efficient. Currently, electronic processes consume unsustainable amounts of energy per operation, which is why future technologies with significantly reduced power consumption are needed. One promising approach is to control electron spin using so-called spintronic devices. However, for this approach to be successful, a deeper understanding of what is happening at the molecular level is needed. To fill that knowledge gap, Mejia Rodriguez’s team is combining physics-based molecular modeling techniques with the latest in machine learning.
The team’s ingenious approach recently achieved more than 450,000 node hours on DOE’s state-of-the-art exascale computing frameworks, Frontier and Aurora at ALCF. The team will use this time to study the electronic structure of metal-based materials that hold promise for spintronic and quantum spin devices. University of Florida experts will input this information to generate machine learning algorithms to further refine their predictions.
“The availability of ExaChem, PNNL’s unique exascale-capable computational chemistry code, will enable us to assess the relevance of high-level computational chemistry methods for modeling next-generation microelectronic devices,” said Mejia Rodriguez. “Complementing these methods with powerful machine learning techniques will advance the exploration of magnetic materials for use in spintronics and quantum information science.”
