Vectorization of the general Monte Carlo classical trajectory program VENUS

The general chemical dynamics computer program VENUS is used to perform classical trajectory simulations for large polyatomic systems, with many atoms and complicated potential energy functions. To simulate an ensemble of many trajectories requires a large amount of CPU time. Since each trajectory is independent, it is possible to parallel process a large set of trajectories instead of processing the trajectories by the conventional sequential approach. This enhances the vectorizability of the VENUS program, since the integration of Hamilton's equations of motion and the gradient evaluation, which comprise 97.8% of the CPU, can each be parallel processed. In this article, the vectorization and ensuing optimization of VENUS on the CRAY‐YMP and IBM‐3090 are presented in terms of both global strategies and technical details. A switching algorithm is designed to enhance the vector performance and to minimize the memory storage. A performance of 140 MFLOPS and a vector/scalar execution rate ratio of 10.6 are observed when this new version of VENUS is used to study the association of CH3 with the H(Ar)12 cluster on the CRAY‐YMP.

[1]  H. Schlegel,et al.  Analytic Function for the H + CH3 CH4 Potential Energy Surface , 1984 .

[2]  Donald G. Truhlar,et al.  Reactive Scattering Cross Sections III: Quasiclassical and Semiclassical Methods , 1979 .

[3]  C. W. Gear,et al.  Numerical initial value problem~ in ordinary differential eqttations , 1971 .

[4]  Donald G. Truhlar,et al.  Strategies and performance norms for efficient utilization of vector pipeline computers as illustrated by the classical mechanical simulation of rotationally inelastic collisions , 1988, Parallel Comput..

[5]  Stephen Wilson,et al.  Methods in Computational Chemistry , 1987 .

[6]  Iain S. Duff,et al.  Architectures and systems , 1989 .

[7]  Bobby G. Sumpter,et al.  Molecular dynamics simulations of polymers: Methods for optimal Fortran programming , 1989 .

[8]  Donald G. Truhlar,et al.  Supercomputer Research in Chemistry and Chemical Engineering , 1987 .

[9]  L. Raff,et al.  Classical Trajectory Methods in Molecular Collisions , 1976 .

[10]  Scott F. Smith,et al.  Theoretical examination of the SN2 reaction involving chloride ion and methyl chloride in the gas phase and aqueous solution , 1985 .

[11]  W. L. Jorgensen,et al.  Effect of hydration on the structure of an SN2 transition state , 1986 .

[12]  Richard B. Bernstein,et al.  Atom - Molecule Collision Theory , 1979 .

[13]  M. D. Pattengill,et al.  Rotational Excitation III: Classical Trajectory Methods , 1979 .

[14]  Michael L. Klein.  COMPUTER SIMULATION STUDIES OF SOLIDS , 1985 .

[15]  D. Fincham,et al.  Recent advances in molecular-dynamics computer simulation , 1985 .

[16]  B. Jönsson,et al.  Vectorizing a general purpose molecular dynamics simulation program , 1986 .

[17]  L. Perera,et al.  Charge localization in negative ion dynamics: Effect on caging of Br−2 in Arn and (CO2)n clusters , 1989 .

[18]  E. Clementi,et al.  Study of the structure of molecular complexes. XII. Structure of liquid water obtained by Monte Carlo simulation with the Hartree–Fock potential corrected by inclusion of dispersion forces , 1975 .

[19]  William H. Miller.  Dynamics of Molecular Collisions , 1976 .

[20]  Wayne A Kraus,et al.  Vectorization of a classical trajectory code on a floating point systems, Inc. Model 164 attached processor , 1986, Journal of computational chemistry.

[21]  D. M. Hirst,et al.  Thermal Rate Constants for H + CH3 CH4 Recombination. II. Comparison of Experiment and Canonical Variational Transition State Theory , 1987 .

[22]  D. C. Rapaport,et al.  Large-scale molecular dynamics simulation using vector and parallel computers , 1988 .