The National Science Foundation (NSF) awarded Georgia Tech and its partners $20 million to build a powerful new supercomputer that uses artificial intelligence (AI) to accelerate scientific breakthroughs.
“Georgia Tech is proud to be one of the nation's leading sources of AI talent and technology that is driving the revolution in our economy,” said Angel Cabrera, president of Georgia Tech. “It's appropriate to be chosen to host this new supercomputer. It supports a new wave of AI-centric innovation across the country. We are grateful to NSF and are excited to get into the work.”
Designed from the ground up for AI, Nexus gives researchers across the country access to advanced computing tools through a simple and user-friendly interface. It supports working in many fields, including climate science, health, aerospace, and robotics.
“Nexus System's novel approach combines support for sustainable scientific services with more traditional, high-performance computing to enable new science and AI workflows that accelerate time to scientific discovery,” says Katie Antypas, Director of the National Science Foundation at Advanced Cyber Infrastructure. “We look forward to adding Nexus to the NSF portfolio of advanced computing capabilities in the research community.”
Nexus SuperComputer – in simple terms
- Built for the Future of Science: Nexus is designed to enhance the most demanding AI research, from treating diseases to understanding how the brain works.
- Burning: The Nexus can close out more than 400 times more operations per second. This is equivalent to everyone in the world, and is continuously performing 50 million calculations per second.
- Large Brain + Memory: Nexus combines the power of AI with high-performance computing, combining 330 trillion bytes of memory to handle complex problems and huge datasets.
- Storage: Nexus features 1 billion bytes of flash storage. This amounts to about 10 billion papers. Stacked, it is a row reaching 500,000 km in height, extending from Earth to the Moon, enough for a third to return.
- Supercharged Connections: The Nexus features lightning connections to move data almost instantly, so researchers don't waste time waiting.
- Researchers Open to the US: Scientists from US institutions can apply for use of Nexus.
AI is rapidly changing the way science researches. Researchers use AI to test large datasets, model complex systems, and ideas faster than ever. However, these tools require powerful computing resources that many institutions have no access to.
This is where Nexus appears. This will ensure cutting-edge AI infrastructure is available not only to Top Tech Hubs, but only scientists across the country.
“The supercomputer helps level the playing field,” says Suresh Marru, principal investigator of the Nexus project and director of Georgia Tech's new Center for AI Science and Engineering (Artisan). “It's designed to make powerful AI tools easier to use, making them available to more researchers in more places.”
“Together with Nexus, Georgia Tech will be joining the league of academic supercomputing centers, a culmination of many years of planning that includes the construction of the state-of-the-art coda data center and the Nexus' “precursor supercompute project.”
Like Nexus, Hive was supported by NSF funding. Both Nexus and The Hive are supported by a partnership between the Georgia Tech Research and Information Technology Units.
Georgia Tech is working with the National Supercomputing Applications (NCSA) at the University of Illinois Urbana-Champaign University to build a Nexus, which operates some of the nation's top academic supercomputers. The two agencies will link systems through new high speed networks to create research infrastructure across the country.
“Nexus is more than a supercomputer. It is a symbol of what is possible when major institutions work together to advance science. “I am proud that my two academic homes are partnering on this project that advances science and society.”
High-tech companies, whose technology powers systems, also play a role.
Georgia Tech will begin building the Nexus, which is expected to be completed in spring 2026 this year. Once Nexus is complete, researchers can apply for access through the NSF review process. Georgia Tech manages the system, provides support and books up to 10% of its own campus research capabilities.
“This is a huge step for Georgia Tech and the scientific community,” said Vivek Sarkar, John P. Imray Dean of Computing. “Nexus helps researchers advance faster on today's toughest issues, opening the door to discoveries we haven't yet imagined.”
