Areas of Research
SPARC projects span multiple disciplines, united by the use of advanced computation, modeling, and AI to address scientific and engineering challenges.
Artificial Intelligence for Science and Engineering
Developing and applying AI foundation models to accelerate discovery in simulation, design, and prediction.
UM Faculty: Venkat Viswanathan, Karthik Duraisamy, Alex Gorodetsky, Venkat Raman
LANL Collaborators: Earl Lawrence, Diane Oyen, Ayan Biswas, Gowri Srinivasan, others.
Hierarchical clustering of logit correlations from a MIST-1.8B variant fine-tuned on scent classification recovers human-interpretable scent relationships by Venkat Viswanathan.
Co-Design of Software and Hardware
Identifying performance bottlenecks and creating new architectures or code strategies to scale up scientific computing.
UM Faculty: Reetu Das, Scott Mahlke
LANL Collaborators: Galen Shipman, Kevin Sheridan, Jered Dominguez-Trujillo
Efficient Memory System Architecture by Reetu Das
Numerical Methods and Algorithms
Advancing methods for multi-scale, multi-physics problems in areas such as astrophysics and high-energy-density physics.
UM Faculty: Alex Gorodetsky, Eric Johnsen, Venkat Raman, Karthik Duraisamy
LANL Collaborators: Josh Dolence, Chad Meyer, Jonah Miller
Irradiation of an overdense clump by two stars by Alex Gorodetsky
Fundamental Physics Modeling
Improving theoretical and computational models across fluid dynamics, plasma physics, turbulence, and condensed matter.
UM Faculty: Aaron Towne, Venkat Raman, Karthik Duraisamy
LANL Collaborators: Daniel Israel
Evolution of the mean energy growth for different horizontal lengths by Aaron Towne
High-Energy-Density Physics (HEDP)
Using simulations and experiments to study matter under extreme conditions, with support from both national and U‑M facilities.
UM Faculty: Carolyn Kuranz, Ryan McBride, Eric Johnsen
LANL Collaborators: Josh Sauppe
Volume fraction and Q-criterion in a three-dimensional case by Carolyn Kurantz
Computational Materials Science
Advancing the theoretical and computational understanding of materials, and developing and deploying next-generation simulation tools.
UM Faculty: Kai Sun, Vikram Gavini, Jacinto Ulloa
LANL Collaborators: Shizeng Lin, Anders Niklasson, Ping Yang, Nitin Daphalapurkar
Compression test results: Normalized vertical displacements with grain breakage by Jacinto Ulloa