GPU-accelerated Monte Carlo simulations of anisotropic Heisenberg ferromagnets

2021/06/21 by Michalis Charilaou, Charilaou, Michalis
Computer Science · Physics and Astronomy · #Computational Physics (physics.comp-ph) #FOS: Physical sciences #Magnetic properties of thin films #Matrix Theory and Algorithms #Mesoscale and Nanoscale Physics (cond-mat.mes-hall) #Statistical Mechanics (cond-mat.stat-mech) #Theoretical and Computational Physics

paper · pdf · doi:10.48550/arxiv.2106.11286

openalex publication_date 2021/06/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The Monte Carlo method is a powerful technique for computing thermodynamic magnetic states of otherwise unsolvable spin Hamiltonians, but the method becomes computationally prohibitive with increasing number of spins and the simulation of real materials and nanostructures is cumbersome. This paper presents the acceleration of Monte Carlo simulations of the three-dimensional anisotropic Heisenberg model on Graphics-Processing Units (GPU). The GPU implementation of the method presented here provides an acceleration of two orders of magnitude over conventional implementations and enables the simulation of large systems, with any crystal lattice, containing up to 10⁸ spins on a single GPU. This offers the possibility to simulate complex structures and devices that are hundreds of nanometers in size in order to compute their magnetic state at finite temperature with atomistic resolution.

Related

• Accelerating micromagnetic and atomistic simulations using multiple GPUs
• The Parallel Monte Carlo Algorithm Implementation on GPU for the Systems with an Ising Hamiltonian under the Condition of a Constant Charge Density
• Critical Phenomena Study of 3D Heisenberg Magnet
• Kernel Fusion in Atomistic Spin Dynamics Simulations on Nvidia GPUs using Tensor Core
• Microcanonical simulated annealing: Massively parallel Monte Carlo simulations with sporadic random-number generation