A simple algorithm to accelerate the computation of non-bonded interactions in cell-based molecular dynamics...

Area of Science:

• Computational physics
• Molecular dynamics simulations

Background:

• Cell lists are essential in molecular dynamics (MD) for calculating inter-atomic potentials and constructing Verlet lists.
• Conventional cell list methods often perform numerous unnecessary interparticle distance calculations, reducing simulation efficiency.

Purpose of the Study:

• To present a novel algorithm that minimizes spurious distance calculations in MD simulations.
• To enhance the computational efficiency of pairwise interaction calculations using cell lists.

Main Methods:

• Developed an algorithm that sorts particles along the cell pair axis.
• Implemented a condition to only calculate interactions between particles if their axial distance is within the cutoff distance.

Main Results:

• The proposed algorithm significantly reduces the number of spurious distance calculations compared to conventional methods.
• The new approach demonstrates superior or comparable efficiency to existing methods, even those using smaller cells.

Conclusions:

• The presented algorithm offers a more efficient method for handling pairwise interactions in MD simulations.
• This optimization is crucial for accelerating large-scale molecular dynamics studies.