HiPerGator the supercomputer has changed the face of engineering research at the University of Florida.
“HiPerGator has fundamentally expanded the scale and ambition of our research,” said Warren Dixon, Ph.D., interim dean of the Herbert Wertheim College of Engineering. “It allows us to test our theories on real, full-scale systems rather than simplified problems, closing the gap between what we can prove on paper and what we can demonstrate in practice.”
As the most powerful and fastest university-owned supercomputer in the nation, HiPerGator enables faculty and students to analyze massive datasets, train sophisticated artificial intelligence models and simulate complex systems with unprecedented speed and scale. That capability has accelerated discovery, strengthened faculty recruitment and given students hands-on experience with the same tools driving industry innovation.
The fourth generation of HiPerGator, unveiled in October, replaced the NVIDIA DGX A100 “Ampere” SuperPOD donated by engineering alumnus and NVIDIA co-founder Chris Malachowsky in 2020 with an NVIDIA DGX B200 SuperPOD.
How is HiPerGator used in the College of Engineering? Here are current projects harnessing the supercomputer to make real-world impacts.
Department of Civil and Coastal Engineering
Eric Du, Ph.D.
Du, the Steve and Wendy Bloom Endowed Professor, is using HiPerGator to push the boundaries of robotics through an NVIDIA-funded project. His team is developing a foundation model for robotic manipulation, using large-scale simulations to train a robotic hand to complete complex tasks like guiding flexible cables through confined spaces, a process known as “wire fishing.”
These tasks are deceptively difficult, requiring the robot to adapt to constantly changing shapes and physical interactions, yet they are essential in real-world applications such as manufacturing, construction and minimally invasive medical procedures. HiPerGator enables the team to run millions of simulated training scenarios, accelerating progress for robots that can operate more effectively in dynamic, real-world environments.
Gary Consolazio, Ph.D.
Professor Consolazio studies how critical infrastructure systems perform under extreme conditions, from vehicle impacts on bridges to barge collisions with flood-control structures. His work is conducted in partnership with organizations like the Florida Department of Transportation and the U.S. Army Corps of Engineers; it relies on large-scale simulations to evaluate the strength, safety and resilience of these systems.
“One of my models might take two days on a normal computer, but on HiPerGator, I can run it in hours,” said Juan Torres, a Ph.D. student in Consolazio’s lab.
That speed allows researchers to test more scenarios, refine designs quickly and make faster, more informed decisions about systems that impact public safety.
Department of Electrical and Computer Engineering
Mark Tehranipoor, Ph.D., and Farimah Farahmandi, Ph.D.
ECE Chair Tehranipoor and Farahmandi, the Walden C. Rhines Endowed Professor for Hardware Security, are using HiPerGator to apply large language models to integrated circuit design and cybersecurity. Their work aims to improve how hardware is designed, protected and regulated in an increasingly complex digital landscape.
Jing Guo, Ph.D.
Associate Professor Guo uses HiPerGator to study the behavior of nanoelectronic devices through multiscale simulations while also developing quantum algorithms. This work bridges traditional computing and emerging quantum technologies, helping to shape the future of small-scale electronics at the smallest scales.
Ivan Ruchkin, Ph.D.
Assistant Professor Ruchkin, the Malachowsky Family Rising Star Assistant Professor, uses HiPerGator to improve the safety and reliability of autonomous systems. His group runs large-scale simulations to test how AI models reason, train machine learning components and evaluate complex systems in high-fidelity virtual environments, which helps reduce risk in real-world deployment.
Janise McNair, Ph.D.
Professor McNair applies advanced AI (deep reinforcement learning) to improve communication networks — like future 6G systems and satellites. Using HiPerGator, her research improves how data moves across complex, multi-tier networks, helping build faster and more efficient global connectivity.
Domenic Forte, Ph.D.
Forte, the Steven A. Yatauro Faculty Fellow and professor, uses HiPerGator to develop AI-driven approaches to hardware security. He focuses on designing “camouflaged” circuits that conceal their true functionality, protecting critical technologies from reverse engineering and cyber threats.
Sandip Ray, Ph.D.
Professor Ray builds digital twin models of human organs, particularly the liver, by combining machine learning with biological data. These models help researchers better understand human health and disease, opening new possibilities for medical research and treatment.
Catia Silva, Ph.D.
Silva, an instructional associate professor, uses HiPerGator to give students hands-on experience with the tools shaping today’s AI industry. In her machine learning courses, students work with large datasets, such as satellite imagery, to detect objects like ships.
“By providing access to HiPerGator in the classroom, students learn more than just concepts; they’re gaining the practical skills and experience they’ll need in industry,” Silva said. “It’s one of the things that really sets our students apart.”
Department of Materials Science and Engineering
Michael Tonks, Ph.D.
For Tonks, MSE’s interim department chair, HiPerGator underpins nearly every aspect of research in materials science. His team uses the system to run large-scale simulations, combined with machine learning, to understand how metals and ceramics evolve over time, particularly in extreme environments.
This work supports a wide range of applications, including the development of next-generation nuclear fuels, modeling how radiation affects materials and predicting corrosion in advanced reactor systems. His group is also contributing to NASA efforts to develop nuclear rocket fuel by simulating the harsh conditions inside propulsion reactors.
By enabling researchers to model material behavior before it can be physically tested, HiPerGator is accelerating the development of safer, more efficient technologies in energy, manufacturing and space exploration.
“HiPerGator,” Dixon concluded, “has expanded the scale and ambition of our research. It allows us to test ideas on real systems, and that’s critical not just for advancing the science, but for preparing students to lead in fields that demand both theoretical depth and computational expertise.”
Department of Mechanical and Aerospace Engineering
Warren Dixon, Ph.D.
In the Nonlinear Controls and Robotics Lab — which Dixon leads — HiPerGator is bridging the gap between mathematical theory and real-world application. His team is building AI that can keep learning while it’s running — rather than being fixed after a single training phase — yet still guarantee it behaves safely and remains stable. HiPerGator lets scientists run the large-scale experiments needed to prove those ideas work, turning concepts once stuck in theory into practical, high-impact solutions.
Sivaramakrishnan Balachandar, Ph.D.
Balachandar, the Newton C. Ebaugh Professor, and his team use the supercomputer to run detailed simulations, such as volcanic eruptions, that show how fluids move and how fires spread. HiPerGator runs massive simulations of natural and engineered systems, such as fluid dynamics and fire behavior, at a scale that would otherwise be out of reach. This allows researchers to uncover insights that traditional computing power might miss.
Johnathon Scheffe, Ph.D.
In energy research, HiPerGator is helping drive more sustainable solutions. Scheffe, associate professor and graduate coordinator, uses the system to simulate heat transfer in solar-driven reactors designed to produce green hydrogen. By modeling complex interactions between fluid flow, heat transfer and radiation, this work supports national efforts to decarbonize industrial processes and advance clean energy technologies.
Doug Spearot, Ph.D.
Spearot, the Newton C. Ebaugh Professor and interim department chair, focuses on how shockwaves damage metallic materials, an issue critical to defense, aerospace and energy systems. His group uses HiPerGator to run large-scale molecular dynamics simulations that capture how materials respond under extreme conditions, where internal damage may not be visible at the surface.
Involving tens of millions of atoms, these simulations allow researchers to study material behavior with detail that would be impossible to achieve experimentally. HiPerGator enables the team to model larger systems over longer timescales, bringing simulations closer to real-world conditions while preparing students to work with advanced computational tools used across industry and national laboratories.