Parallel Markov chain Monte Carlo Simulation by Pre-Fetching

Some existing methods for parallelization of Markov chain Monte Carlo algorithms are discussed, and a new “pre-fetching” algorithm is proposed to parallelize generation of a single chain.

Abstract

In recent years, parallel processing has become widely available to researchers. It can be applied in an obvious way in the context of Monte Carlo simulation, but techniques for ``parallelizing'' Markov chain Monte Carlo (MCMC) algorithms are not so obvious, apart from the natural approach of generating multiple chains in parallel. While generation of parallel chains is generally the easiest approach, in cases where burn-in is a serious problem, it is often desirable to use parallelization to speed up generation of a single chain. This paper briefly discusses some existing methods for parallelization of MCMC algorithms, and proposes a new ``pre-fetching'' algorithm to parallelize generation of a single chain.