Calculating Complex Interactions in Molecular Dynamics Simulations Employing Lagrangian Particle Tracking Schemes

Abstract

In this report, we describe some general features of Lagrangian particle tracking and the method of table look-up that are important for calculating many different types of interactions of arbitrary complexity in large molecular dynamics simulations. Based on our experience using these approaches and state-of-the-art of computer technology, we state formally those properties of Lagrangian tracking and table look-up that are important to the design of optimal programming structures. We examine the relationship between Lagrangian indexing and list indexing of data and how this influences the complexity of programming structures for effecting vector operations. Included in this report are criteria for the efficient use of Lagrangian particle tracking schemes in molecular dynamics simulations.