Finite difference micromagnetic solvers with Object Oriented Micromagnetic Framework on graphics processing units

Published

November 23, 2015

Author(s)

Michael J. Donahue, Sidi Fu, Weilong Cui, Matthew Hu, Ruinan Chang, Vitaliy Lomakin

Abstract

A micromagnetic solver with Finite Difference method on Graphics Processing Unit (GPU) and its integration with the Object Oriented MicroMagnetic Framework (OOMMF) are presented. Two approaches for computing the magnetostatic field accelerated by Fast Fourier Transform (FFT) are implemented. The first approach, referred to as the tensor approach, is based on the tensor spatial convolution to directly compute magnetostatic field from magnetic moments. The second approach, referred to as the scalar potential approach, is a novel technique that uses differential operator evaluation through finite differences (divergence for magnetic charge and gradient for magnetostatic field) and spatial convolution for magnetic scalar potential. Comparisons of implementation details, speed, memory consumption and accuracy are provided. The GPU implementation of OOMMF shows up to 25x GPU-CPU speed-up.

Citation

Journal of Magnetism and Magnetic Materials

Pub Type

Journals

Download Paper

https://doi.org/10.1109/TMAG.2015.2503262

Local Download

Keywords

micromagnetics, graphics processing unit, fast Fourier transform, demagnetization field

Citation

Donahue, M. , , S. , Cui, W. , Hu, M. , Chang, R. and Lomakin, V. (2015), Finite difference micromagnetic solvers with Object Oriented Micromagnetic Framework on graphics processing units, Journal of Magnetism and Magnetic Materials, [online], https://doi.org/10.1109/TMAG.2015.2503262 (Accessed November 23, 2024)