The U.S. Department of Energy’s (DOE) Argonne National Laboratory hosted a ribbon-cutting ceremony last week to celebrate its new Aurora exascale computer, marking a major milestone for AI-powered science. The event brought together DOE leaders, researchers, and partners from Intel and HPE to recognize the collaborative effort behind Aurora and its potential to transform scientific discovery.
“Aurora is a powerful example of what American science and innovation can deliver,” said U.S. Secretary of Energy Chris Wright. “As global competition accelerates, systems like Aurora give the United States a decisive edge in artificial intelligence, scientific discovery, and national security — fields where we can’t afford to fall behind.”
Deployed for science earlier this year, Aurora is one of the most powerful computing systems ever built. The machine is designed to accelerate breakthroughs in science and engineering by combining simulation, AI and data analysis capabilities at an unprecedented scale.
“Aurora is more than a technological achievement — it is a strategic national asset,” said Argonne Laboratory Director Paul Kearns. “By harnessing the power of exascale computing and AI, our new system gives researchers across the country a critical tool for tackling big questions in energy, medicine and materials. It reflects our commitment to building dynamic ecosystems that bring together technology, talent and purpose to advance discovery and improve lives through science.”
Aurora is one of three DOE supercomputers to surpass the exascale threshold, capable of performing more than a quintillion calculations per second. Along with Frontier at Oak Ridge National Laboratory and El Capitan at Lawrence Livermore National Laboratory, the systems occupy the top three spots on both the TOP500 list of the world’s fastest supercomputers and the HPL-MxP benchmark for AI performance. These results highlight DOE’s global leadership in supercomputing and AI-driven research.
Developed in collaboration with Intel and HPE, Aurora is equipped with 63,744 graphics processing units (GPUs), making it one of the largest GPU-powered supercomputers in the world.
“Aurora’s latest achievement reflects the scale, ambition and deep technical collaboration between Intel, HPE and Argonne National Laboratory,” said Chris George, president of Intel government technologies. “Powered by Intel’s unified hardware and software architecture, Aurora brings together AI and high performance computing to accelerate scientific discovery. It’s a powerful example of what public-private collaboration can deliver and a foundation for the next generation of systems that will push the boundaries of innovation.”
Available to researchers across the nation, Argonne’s Aurora system provides advanced AI, simulation and data analysis capabilities to accelerate scientific discoveries and innovation. (Image by Argonne National Laboratory.)
Occupying 10,000 square feet (roughly the size of two basketball courts), the massive system also features advanced water-cooling infrastructure and one of the most expansive interconnects ever built, with over 300 miles of networking cables and 84,992 network endpoints.
“The Aurora exascale supercomputer represents a new era for science — its breakthrough engineering and unprecedented performance powers a next generation of modeling, simulation and AI,” said Trish Damkroger, senior vice president and general manager of HPC & AI Infrastructure Solutions at HPE. “We are proud to have partnered with Intel to build Aurora and, together, play a significant role in a strong public-private partnership with the U.S. Department of Energy and Argonne to bring a bold vision to life for the scientific community.”
Aurora is housed at the Argonne Leadership Computing Facility, a DOE Office of Science user facility available to researchers from across the U.S. and around the world.
Aurora in Action: Accelerating Science with AI and Simulation
Aurora is already enabling breakthroughs across a wide range of scientific fields. In biology and medicine, researchers are using AI and simulation on Aurora to predict how viruses evolve, improve cancer treatments and map the neural connections in the brain. By combining large-scale AI models with physics-based methods, Aurora is enabling research that aims to accelerate drug discovery, enhance global health preparedness and deepen our understanding of complex biological systems.
In aerospace engineering, Aurora is helping scientists better understand how air flows around aircraft during flight. High-fidelity exascale simulations allow them to capture complex phenomena such as turbulence, shock waves and the lift generated during takeoff and landing. Teams are also using Aurora to explore next-generation propulsion systems by modeling airflow around advanced engine and wing configurations. The insights gained from these studies help improve predictive models, thereby reducing the need for physical testing and accelerating the design of quieter, more efficient aircraft.
Aurora is also playing a key role in advancing the pursuit of fusion energy. Scientists are simulating the extreme conditions inside reactors like the ITER international fusion energy project, tracking how heat, particles and impurities move through plasma and interact with reactor walls. Alongside these simulations, researchers are using AI to predict and control the behavior of energetic particles, a key step toward managing the fusion process. With Aurora’s advanced capabilities, these simulation and AI approaches are helping enable breakthroughs needed to move fusion closer to reality.
In quantum computing, Aurora’s exascale power allows researchers to run large-scale simulations that support and validate quantum experiments. Scientists are using the system to confirm the reliability of randomness generated by quantum devices — an important feature for secure communication that is difficult to verify. They are also testing quantum algorithms designed to solve complex problems in molecular design, helping explore how quantum and classical computing can work together in future scientific applications.
By delivering faster, more accurate insights across research fields, Aurora is powering a new era of science driven by AI, simulation, data and collaboration. The breakthroughs it enables are expected to have far-reaching impacts, shaping new technologies and tools that enhance everyday life.
This article was originally published by Argonne National Laboratory and is reprinted with permission.