Computational Physics on Connection Machine Supercomputers
Developments in computer architectures are changing the way we do computational physics. Data-parallel programming languages eliminate unnecessary serialization, which obscures the parallelism in nature, and allow scientists to program computers at higher levels. Massively parallel hardware designs allow both higher performance and better performance-per-price. In this talk, I will discuss data-parallel programming, Connection Machine (CMTM) design, and selected applications on Connection Machine supercomputers from computational condensed-matter physics.
KeywordsComputer Architecture Rectangular Grid Processing Node Code Fragment Data Parallelism
Unable to display preview. Download preview PDF.
- 1.See also Connection Machine Model CM-2 Technical Summary, Version 6.0, November 1990, Thinking Machines Corp., Cambridge, Mass.Google Scholar
- 2.See also B.M. Boghosian, Computers in Physics 4, 14 (1990).Google Scholar
- 5.J. D. Joannopoulos, K. Brommer, B. E. Larson, M. Nccdcls, to be published.Google Scholar
- 6.J.T. Camm el, D.K. Campbell, L. Loh, S. Mazumdar, S.N. Dixit, to appear in J. Materials Research.Google Scholar