Regular lattice and small-world spin model simulations using CUDA and GPUs

Data-parallel accelerator devices such as Graphical Processing Units (GPUs) are providing dramatic performance improvements over evenmulti-coreCPUs for lattice-oriented applications in computational physics. Models such as the Ising and Potts models continue to play a role in investigating phase transitions on smallworld and scale-free graph structures. These models are particularly well-suited to the performance gains possible using GPUs and relatively high-level device programming languages such as NVIDIA'sComputeUnified Device Architecture (CUDA).We report on algorithms andCUDAdata-parallel programming techniques for implementingMetropolis Monte Carlo updates for the Isingmodel using bit-packing storage, and adjacency neighbour lists for various graph structures in addition to regular hypercubic lattices. We report on parallel performance gains and also memory and performance tradeoffs using GPU/CPU and algorithmic combinations