Vectorized Monte Carlo molecular aerodynamics simulation of the Rayleigh problem

Abstract

A fully vectorized version of a Monte Carlo algorithm applied to molecular aerodynamics is presented. The algorithm is applied to the one-dimensional Rayleigh problem, an impulsively accelerated, heated flat plate, and is implemented on the CYBER 205. Issues relating to the details of vectorization, including the use of bit-type vectors, the maintaining of long vector lengths, and vector gather/scatter use, are discussed. Timings as functions of problem size are presented for various facets of the algorithm, and the problem of the "tail" due to stragglers is examined and quantified. Asymptotic speedup factors due to vectorization in the neighborhood of 7 to 8 are reported, depending upon problem parameters. The results indicate that, under certain circumstances, the problem performs well on vector architectures.

Parallelization of direct simulation Monte Carlo method combined with monotonic Lagrangian grid

“…29 " 31 The DSMC one-dimensional solutions of this problem have been validated by comparing them with both Bhatnagar-Gross-Krook and analytical solutions. 1 Variations of the Rayleigh problem were used to validate two-dimensional versions of our parallelized DSMC code.…”

Vectorized algorithm for multidimensional Monte Carlo integration on modern GPU, CPU and MIC architectures

“…Using parallel computers can accelerate the performance of programs that use such methods in science, technology or financial applications [3,20,22,24]. Monte Carlo methods are also very attractive for solving multidimensional integration problems [1].…”