Frameworks, Algorithms and Scalable Technologies
for Mathematics (FASTMath) SciDAC-5 Institute
The FASTMath SciDAC-5 Institute develops and deploys scalable mathematical algorithms, AI/ML capabilities, and software tools for reliable simulation and analysis of complex physical phenomena and collaborates with application scientists to ensure the usefulness and applicability of FASTMath technologies.
Objectives of the SciDAC-5 FASTMath Institute
develop robust mathematical techniques, numerical algorithms, and AI/ML capabilities to reliably address the challenges of large-scale simulation and analysis of complex physical phenomena;
work closely with domain scientists to leverage our mathematical and AI/ML expertise and deploy our software in large-scale modeling and simulation codes;
deliver highly performant software with strong software engineering to run efficiently and scalably on current and next-generation advanced computer architectures at the DOE Office of Science’s three major computing facilities; and
build and support the broader computational mathematics and computational science communities across the DOE complex.
FASTMath has eight thrusts that broadly span the mathematical methodologies required by SciDAC community. AI/ML is a cross-cutting theme, with synergistic activities occurring across the Institute.
As SciDAC Partnerships solicitations are released by DOE, a summary of FASTMath resources available to proposal teams can be found at this page https://scidac5-fastmath.lbl.gov/resources-for-scidac-partnerships, where meetings with the FASTMath Institute can be requested.
Questions?
Contact FASTMath at listserv.llnl.gov to get more information on the project.
The FASTMath Institute is funded by the Scientific Discovery through Advanced Computing (SciDAC) Program under the Office of Science at the U.S. Department of Energy.