A Fast Algorithm for Massively Parallel, Long-Term, Simulation of Complex Molecular Dynamics Systems

The advances in theory and computing technology over the last decade have led to enormous progress in applying atomistic molecular dynamics (MD) methods to the characterization, prediction, and design of chemical, biological, and material systems. MD simulation of very large systems are currently being performed by using massively parallel architectures. However, there is a fundamental difficulty with time and length scales that obstructs these MD methodologies for application to some of the most important problems. In fact, with the current technologies it is routine to consider the MD of large scale systems for a time scale of nanosecond. The problem is that many important applications require a time scale of a microsecond or more. In order to achieve such a significant improvement in the computational efficiency, in addition to massively parallel platforms, new algorithms are needed to allow efficient long-term MD simulation.