Mixed precision and mixed model numerical methods that are computationally and energy efficient
Abstract
Our goal in this proposal is to develop computationally- and energy-efficient numerical algorithms for time dependent PDEs, using a mixed-precision, mixed-model approach. This multi-faceted and systematic advancement of power-efficient algorithms for numerical PDEs will enable the development of scientific codes fully capitalizing on the novel cost-efficient hardware. We are therefore making an important step towards green computing and reduced dependence on fossil fuels and the associated pollution.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 22, 2024
- Source ID
- FA95502310037
Entities
People
- Sigal Gottlieb
Organizations
- Air Force Office of Scientific Research
- United States Air Force
- University of Massachusetts