Mathematics and Computer Science
Exascale Computing in the MCS Division
Providing breakthrough tools and technologies to exploit exascale computing platforms for scientific discovery
Researchers in Argonne’s Mathematics and Computer Science Division are involved in several new projects involving exascale computing.
Algorithms and Software
- X-CELLENT: X-Compiler Extending LLVM for Enhanced Natural Translation – Paul Hovland
- Automatic Generation of Algorithms for High-Speed Reliable Lossy Compression – Sheng Di
- Tensor-Compressed Sustainable Pre-Training of Extreme-Scale Foundation Models – Franck Cappello
- Scalable Learning and Optimization for Secure and Economic Grid Operations – Emil Constantinescu
- DataStates: Scalable Versioning for Scientific Data – Bogdan Nicolae
Stewardship
- S4PST (Stewardship for Programming Systems and Tools) -- Michel Schanen
- Next-Generation Ecosystems for Scientific Computing: Harnessing Community, Software, and AI for Mission-Driven Team Science – Lois Curfman McInnes
- PESO: Partnering for Scientific Software Ecosystem Stewardship Opportunities– Lois Curfman McInnes
- STEP: Software Tools Ecosystem Project – Philip Carns
Institutes and Laboratories
- FASTMath: Frameworks, Algorithms and Scalable Technologies for Mathematics (a SciDAC-5 Institute) – Todd Munson
- RAPIDS2: A SciDAC Institute for Computer Science, Data, and Artificial Intelligence – Robert Ross
- JLESC: Joint Laboratory on Extreme Scale Computing – Franck Cappello
Application Development
- MPAS-O/ROMS Comparison, Nesting, and Coupling for Improved Representation and Parameterization of Coastal and Submesoscale Ocean Processes in E3SM – Iulian Grindeanu
- Femtoscale Imaging of Nuclei using Exascale Platforms – Emil Constaninescu
- SciDAC-5 FES Partnerships – Jeffrey Larson
- Modeling of DED Process with OpenFOAM Using GPU-accelerated Solvers and Machine Learning Trained Source Terms – Junchao Zhang
- Fundamental Nuclear Physics at the Exascale and Beyond – Robert Ross
Note on older projects: MCS was heavily involved in the DOE Exascale Computing Project, which began in 2016 and ended in 2024. For a list of our activities under that project, see the website.
