1096-987X/asset/JCC_centre.gif?v=1&s=b0d6b2f567f5e92bfd33499dcef2c15d54c9375d)
A massively multicore parallelization of the Kohn-Sham energy gradients
Article first published online: 1 FEB 2010
DOI: 10.1002/jcc.21485
Copyright © 2010 Wiley Periodicals, Inc.
1096-987X/asset/cover.gif?v=1&s=4429aac2462ebd499c13b3d7fe983679c5767778 "Journal of Computational Chemistry")
Additional Information
How to Cite
Brown, P., Woods, C. J., McIntosh-Smith, S. and Manby, F. R. (2010), A massively multicore parallelization of the Kohn-Sham energy gradients. J. Comput. Chem., 31: 2008–2013. doi: 10.1002/jcc.21485
Publication History
- Issue published online: 21 MAY 2010
- Article first published online: 1 FEB 2010
- Manuscript Accepted: 16 NOV 2009
- Manuscript Received: 1 OCT 2009
Funded by
- School of Chemistry at the University of Bristol
- Clear Speed Technology plc
- EPSRC. Grant Number: EP/F010516/1
- Abstract
- Article
- References
- Cited By
Keywords:
- density functional theory;
- gradients;
- parallel;
- high performance computing;
- GPU;
- electronic structure theory
Abstract
In a previous article [Brown et al., J Chem Theory Comput 2009, 4, 1620], we described a quadrature-based formulation of the Kohn-Sham Coulomb problem that allows for efficient parallelization over thousands of small processor cores. Here, we present the analytic gradients of this modified Kohn-Sham scheme, and describe the parallel implementation of the gradients on a numerical accelerator architecture. We demonstrate an order-of-magnitude acceleration for the combined energy and gradient calculation over a conventional single-core implementation. © 2010 Wiley Periodicals, Inc. J Comput Chem, 2010