Premium
A coarse‐grain parallel implementation of the direct SCF method
Author(s) -
Lüthi Hans P.,
Mertz John E.,
Feyereisen Martin W.,
Almlöf Jan E.
Publication year - 1992
Publication title -
journal of computational chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.907
H-Index - 188
eISSN - 1096-987X
pISSN - 0192-8651
DOI - 10.1002/jcc.540130207
Subject(s) - mimd , speedup , parallel computing , computer science , computation , overhead (engineering) , scheme (mathematics) , computational science , algorithm , mathematics , mathematical analysis , operating system
A coarse‐grain, parallel approach to direct Hartree–Fock calculations is presented and discussed. The suggested scheme allows for a near asymptotic speedup involving a very low parallelization overhead without compromising the vector performance of vector‐parallel architectures. A shared‐memory MIMD implementation, for which very high speeds of computation have been achieved, is discussed in detail.