Case Studies

The ALCF user community continues to advance scientific computing through large-scale simulations, AI-driven methods, and data-intensive research. These case studies highlight a range of projects using ALCF’s HPC and AI resources to break new ground in science and engineering.

Domain Chemistry
Systems Polaris, Theta
PI Wei Jiang, Argonne National Laboratory

Microscopic Insight into Transport Properties of Li-Battery Electrolytes

Molecular dynamics simulation of electrolyte/electrode interaction

Domain Computer Science
Systems Polaris, Aurora
PI Aiichiro Nakano, University of Southern California

Accelerating Light-Matter Dynamics Simulations with DCMESH

Graph showing speedups with the DCMESH code

Domain Computer Science
Systems Polaris
PI Michael Prince, Argonne National Laboratory

Opportunities for Retrieval and Tool-Augmented Large Language Models in Scientific Facilities

Overview of CALMS

Domain Computer Science
Systems Polaris
PI Christopher Graziul, University of Chicago

Speech Recognition for Analysis of Police Radio Communication

Speech Recognition for Analysis of Police Radio Communication

Domain Computer Science
Systems Polaris
PI Victor Mateevitsi, Argonne National Laboratory

Bridging Gaps in Simulation Analysis through a General Purpose, Bidirectional Steering Interface with Ascent

An example flow field for a real tree case from Lidar data

Domain Computer Science
Systems Aurora
PI Thomas Applencourt, Argonne National Laboratory

Ponte Vecchio Across the Atlantic: Single-Node Benchmarking of Two Intel GPU Systems

Figures of merit for Aurora and Dawn

Domain Computer Science
Systems Polaris, Aurora
PI Allan Grosvenor, MSBAI

Autonomy for DOE Simulations

GURU solution-adaptive mesh refinement example

Domain Engineering
Systems Polaris
PI Shivam Barwey, Argonne National Laboratory
Pinaki Pal, Argonne National Laboratory

Mesh-Based Super-Resolution of Fluid Flows with Multiscale Graph Neural Networks

Visualizations of vorticity contours

Systems Aurora
PI Rick Stevens, Argonne National Laboratory

AuroraGPT: A Large-Scale Foundation Model for Advancing Science

Systems Aurora
PI Kenneth Jansen, University of Colorado

PHASTA: Machine Learning-Guided Computational Fluid Dynamics at Extreme Scales