Difficulties with Multiple Time Stepping and the Fast Multipole Algorithm in Molecular Dynamics

Tuesday, January 21, 1997 - 11:00am - 12:00pm
Robert Skeel (University of Illinois at Urbana-Champaign)
Numerical experiments are performed on a 36,000-atom protein--DNA--water simulation to ascertain the effectiveness of two devices for reducing the time spent computing long-range electrostatics interactions. It is shown for Verlet-I/r-RESPA multiple time stepping, which is based on approximating long-range forces as widely separated impulses, that a long time step of 5fs results in a dramatic energy drift and that this is reduced by using a yet larger long time step. Also, it is shown that the use of as many as eight terms in a fast multipole algorithm approximation to long-range electrostatics still fails to prevent significant energy drift even though 5 digits of accuracy is obtained.

This is joint work with T.C. Bishop and K. Schulten.