Premium
Recent performance improvements to the DFT and TDDFT in GAMESS
Author(s) -
Lasinski Michael E.,
Romero Nichols A.,
Brown Shawn T.,
Blaudeau JeanPhilippe
Publication year - 2012
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.22890
Subject(s) - time dependent density functional theory , density functional theory , computer science , computational science , software , supercomputer , chemistry , computational chemistry , parallel computing , programming language
The general atomic and molecular electronic structure system (GAMESS) is a quantum chemistry package used in the first‐principles modeling of complex molecular systems using density functional theory (DFT) as well as a number of other post‐Hartree‐Fock methods. Both DFT and time‐dependent DFT (TDDFT) are of particular interest to the materials modeling community. Millions of CPU hours per year are expended by GAMESS calculations on high‐performance computing systems; any substantial reduction in the time‐to‐solution for these calculations represents a significant saving in CPU hours. As part of this work, three areas for improvement were identified: (1) the exchange‐correlation (XC) integration grid, (2) profiling and optimization of the DFT code, and (3) TDDFT parallelization. We summarize the work performed in these task areas and present the resulting performance improvement. These software enhancements are available in 12JAN2009R3 or later versions of GAMESS. © 2012 Wiley Periodicals, Inc. J Comput Chem, 2012