Performance Portable Monte Carlo Particle Transport on Intel, NVIDIA, and AMD GPUs

TL;DR

This paper demonstrates that OpenMC, a Monte Carlo particle transport application, achieves performance portability across AMD, NVIDIA, and Intel GPUs on supercomputers, using the OpenMP target offloading model, and compares its performance to CPU versions and other applications.

Contribution

It is among the first to show large scientific application performance at scale on Intel Ponte Vecchio GPUs and across multiple GPU vendors using OpenMP offloading.

Findings

OpenMC performs efficiently on AMD, NVIDIA, and Intel GPUs.

OpenMC outperforms CPU versions and other Monte Carlo applications.

First large-scale scientific simulation results on Intel Ponte Vecchio GPUs.

Abstract

OpenMC is an open source Monte Carlo neutral particle transport application that has recently been ported to GPU using the OpenMP target offloading model. We examine the performance of OpenMC at scale on the Frontier, Polaris, and Aurora supercomputers, demonstrating that performance portability has been achieved by OpenMC across all three major GPU vendors (AMD, NVIDIA, and Intel). OpenMC's GPU performance is compared to both the traditional CPU-based version of OpenMC as well as several other state-of-the-art CPU-based Monte Carlo particle transport applications. We also provide historical context by analyzing OpenMC's performance on several legacy GPU and CPU architectures. This work includes some of the first published results for a scientific simulation application at scale on a supercomputer featuring Intel's Max series "Ponte Vecchio" GPUs. It is also one of the first demonstrations of a large scientific production application using the OpenMP target offloading model to achieve high performance on all three major GPU platforms.