Chemists continue to explore the use of very large computations to perform simulations that describe the molecular level physics of critical challenges in science. In this paper, the Dynamical Nucleation Theory Monte Carlo (DNTMC) model - a model for determining molecular scale nucleation rate constants - and its parallel capabilities are described. The potential for bottlenecks and the challenges to running on future petascale or larger resources are delineated. A "master-slave" solution is proposed to scale to the petascale and will be developed in the NWChem software. In addition, mathematical and data analysis challenges are also described. This work was supported by the U.S. Department of Energy's (DOE) Office of Basic Energy Sciences, Chemical Sciences program. The Pacific Northwest National Laboratory is operated by Battelle for DOE.
Revised: April 7, 2011 |
Published: July 14, 2008
Citation
Windus T.L., S.M. Kathmann, and L.D. Crosby. 2008.High performance computations using dynamical nucleation theory.Journal of Physics: Conference Series 125.PNNL-SA-61386.doi:10.1088/1742-6596/125/1/012017