Large-scale Ferrofluid Simulations on Graphics Processing Units

TL;DR

This paper introduces a GPU-optimized molecular-dynamics simulation method for ferrofluids, achieving significant speed-ups over traditional CPU and GPU algorithms while maintaining accuracy through comparison with theoretical models.

Contribution

A novel GPU-oriented modification of the Barnes-Hut algorithm for large-scale ferrofluid simulations, enabling efficient and accurate modeling of one million particles.

Findings

Four orders of magnitude speed-up over CPU All-Pairs algorithm

Two orders of magnitude speed-up over GPU-optimized All-Pairs

Simulation results agree with theoretical predictions

Abstract

We present an approach to molecular-dynamics simulations of ferrofluids on graphics processing units (GPUs). Our numerical scheme is based on a GPU-oriented modification of the Barnes-Hut (BH) algorithm designed to increase the parallelism of computations. For an ensemble consisting of one million of ferromagnetic particles, the performance of the proposed algorithm on a Tesla M2050 GPU demonstrated a computational-time speed-up of four order of magnitude compared to the performance of the sequential All-Pairs (AP) algorithm on a single-core CPU, and two order of magnitude compared to the performance of the optimized AP algorithm on the GPU. The accuracy of the scheme is corroborated by comparing the results of numerical simulations with theoretical predictions.