Adaptive Winograd's matrix multiplications
Abstract
Modern architectures have complex memory hierarchies and increasing parallelism (e.g., multicores). These features make achieving and maintaining good performance across rapidly changing architectures increasingly difficult. Performance has become a complex tradeoff, not just a simple matter of counting cost of simple CPU operations.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Mar 01, 2009
- Source ID
- 10.1145/1486525.1486528
Entities
People
- Alexandru Nicolau
- Paolo D'alberto
Organizations
- Defense Advanced Research Projects Agency
- Jerry and David's guide to the World Wide Web
- University of California