|
Abstract : |
Abstract. We describe the results of a preliminary port of a large molecular dynamics (MD) code, gromos [15], to a distributed--memory parallel computer. The objectives of this study were three fold. First, we wanted to assess the suitability of our software tools [14] for porting existing Fortran codes (dusty decks). This involved developing Fortran versions of previously developed parallel extensions to the C programming language. Secondly, we wanted to be able to quantify various components of an MD simulation with respect to communication and computation costs. Here the objective was to have timing data that would aid in the design of a scalable code that could execute efficiently on a large number of processors. Finally, we wanted to see if execution speeds comparable to a Cray could be achieved easily with a modest number of current--generation nodes. Such performance seemed feasible based on a rough assessment of current processor performance and communication speed. One of our principal conclusions is that we were able to achieve performance on a standard benchmark problem comparable to reported performance of a Cray--2 using between 8 and 16 processors on an iPSC/860. Moreover, this was done with a minimal amount of coding and debugging, due primarily to the use of our software tools. 1. Description of MD algorithms. The basic model of molecular dynamics [1, 5, 12] involves only classical mechanics, although the forces, F, between atoms are more complicated than the gravitational or electrostatic forces encountered in basic mechanics. The model reduces 1 to a system of ordinary differential equations for the positions, x, of the atoms which express essentially that F = m, |
