Asynchronous quasi-monte Carlo methods

摘要

We present an asynchronous algorithm based on Quasi-Monte Carlo methods for the computation of multivariate integrals. Randomized Korobov and Richtmyer sequences are applied for problems of moderately high to high dimensions. Sobol rules are a candidate for dealing with certain singular behavior. Timing results are obtained on MPI for a class of integrals from computational finance, and yield good speedup on a network of workstations.