Speedup of Micromagnetic Simulations with C++ AMP on Graphics Processing Units

A finite-difference micromagnetic solver called Grace uses C++ Accelerated Massive Parallelism (C++ AMP). The high-speed performance of a single GPU is compared against a typical CPU-based solver. The speedup of GPU to CPU is shown to be two orders of magnitude for problems with larger sizes. This solver can run on GPUs from various hardware vendors, such as Nvidia, AMD, and Intel, regardless of whether it is a dedicated or integrated graphics processor. The Web extra for this article includes appendices of code from the author's original work.