Abstract

The design, validation and performance of Grace, a GPU-accelerated micromagnetic simulation software, are presented. The software adopts C+ + Accelerated Massive Parallelism (C+ + AMP) so that it runs on GPUs from various hardware vendors including NVidia, AMD and Intel. At large simulation scales, up to two orders of magnitude of speedup factor is observed, compared to CPU-based micromagnetic simulation software OOMMF. The software can run on high-end professional GPUs as well as budget personal laptops, and is free to download.

Highlights

  • There have been several research groups working on the GPU-accelerated micromagnetic simulation software.[1,2,3,4,5] The goal is to harness the parallel computing power of GPUs to speed up the micromagnetic simulation, which is predominated by finite difference or finite element calculations that can be parallelized

  • The software adopts C++ Accelerated Massive Parallelism (C++ AMP) so that it runs on GPUs from various hardware vendors including NVidia, AMD and Intel

  • NVidia’s Compute Unified Device Architecture (CUDA) is used exclusively in the previously mentioned works, prohibiting the use of other widely available GPU platforms provided by vendors including AMD and Intel, such as AMD Radeon, FirePro, and Intel Xeon Phi

Read more

Summary

INTRODUCTION

There have been several research groups working on the GPU-accelerated micromagnetic simulation software.[1,2,3,4,5] The goal is to harness the parallel computing power of GPUs to speed up the micromagnetic simulation, which is predominated by finite difference or finite element calculations that can be parallelized. NVidia’s Compute Unified Device Architecture (CUDA) is used exclusively in the previously mentioned works, prohibiting the use of other widely available GPU platforms provided by vendors including AMD and Intel, such as AMD Radeon, FirePro, and Intel Xeon Phi. In this work, Grace, a cross-hardware platform GPU-accelerated micromagnetic simulation software based on C++ AMP, is presented, which speeds up the simulation time by up to two orders of magnitudes, as compared CPU-based micromagnetic simulation software OOMMF.

DESIGN
VALIDATION
PERFORMANCE
CONCLUSION
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.