Verlet list

A Verlet list (named after Loup Verlet) is a data structure in molecular dynamics simulations to efficiently maintain a list of all particles within a given cut-off distance of each other.[1]

This method may easily be applied to Monte Carlo simulations. For short-range interactions, a cut-off radius is typically used, beyond which particle interactions are considered "close enough" to zero to be safely ignored. For each particle, a Verlet list is constructed that lists all other particles within the potential cut-off distance, plus some extra distance so that the list may be used for several consecutive Monte Carlo "sweeps" before being updated. If we wish to use the same Verlet list n times before updating, then the cut-off distance for inclusion in the Verlet list should be R_c + 2nd, where R_c is the cut-off distance of the potential, and d is the maximum Monte Carlo step of a single particle. Thus, we will spend of order N^2 time to compute the Verlet lists (N is the total number of particles), but are rewarded with n Monte Carlo "sweeps" of order Nn^2 (instead of NN). Optimizing our choice of n, it can be shown that the O(N^2) problem of Monte Carlo sweeps has been converted to an O(N^{5/3}) problem by using Verlet lists.

Using cell lists to identify the nearest neighbors in O(N) further reduces the computational cost.

See also

References

  1. Verlet, L. (1967). "Computer 'experiments' on classical fluids. I. Thermodynamical properties of Lennard-Jones molecules". Phys. Rev. 159: 98–103. doi:10.1103/physrev.159.98.

External links


This article is issued from Wikipedia - version of the Sunday, January 18, 2015. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.