In addition to his role as a professor, UT Computer Science’s Robert van de Geijn is the Director of the ICES’ Science of High-Performance Computing (SHPC) Group. Van de Geijn, drawing upon expertise from years of research in linear algebra, high-performance computing, parallel computing, and formal derivation of algorithms, is leading the SHPC Group, in collaboration with the Intel Parallel Computing Centers (Intel PCC), to enable large computer code performance increases while maintaining code portability. Both groups will develop novel algorithms and runtime systems for performing tensor contractions on multi- and many-core systems. The algorithms are a critical computational step in high-accuracy quantum chemical calculations necessary to describe fully the structure, kinetics, spectroscopy and other physics of atoms and molecules.
The project will develop novel algorithms and runtime systems for performing tensor contractions on multi- and many-core systems. The algorithms are a critical computational step in high-accuracy quantum chemical calculations necessary to describe fully the structure, kinetics, spectroscopy and other physics of atoms and molecules.
“We hope to leverage several key Intel technologies as we strive to increase efficiency and scalability of quantum chemistry calculations on current and future Intel architectures,” says Devin Matthews, the Arnold O. Beckman Postdoctoral Fellow developing the algorithms under Professor Robert van de Geijn.
Van de Geijn’s SHPC Group focuses on the knowledge that underlies scientific software libraries and makes it systematic. On the one hand, this has facilitated the software architecture of such libraries. On the other hand, this allows for making much of the development of such software mechanical via goal-oriented programming techniques from formal methods and Design-by-Transformation from software engineering.
The groundbreaking research being done by both the van de Geijn’s ICES SHPC Group and the Intel PCC contributes to the foundations of the field. It has also created a new dense linear algebra software stack that can nimbly respond to the rapid changes that computer architecture is experiencing. These software artifacts are available under open software license.