A scalable low discrepancy point generator for parallel computing

Abstract

The Monte Carlo (MC) method is a simple but effective way to perform simulations involving complicated or multivariate functions. The Quasi-Monte Carlo (QMC) method is similar but replaces independent and identically distributed (i.i.d.) random points by low discrepancy points. Low discrepancy points are regularly distributed points that may be deterministic or randomized. The digital net is a kind of low discrepancy point set that is generated by number theoretical methods. A software library for low discrepancy point generation has been developed. It is thread-safe and supports MPI for parallel computation. A numerical example from physics is shown. © Springer-Verlag Berlin Heidelberg 2004.