Distributed OpenMP Offloading of OpenMC on Intel GPU MAX Accelerators (2403.02735v2)
Abstract: Monte Carlo (MC) simulations play a pivotal role in diverse scientific and engineering domains, with applications ranging from nuclear physics to materials science. Harnessing the computational power of high-performance computing (HPC) systems, especially Graphics Processing Units (GPUs), has become essential for accelerating MC simulations. This paper focuses on the adaptation and optimization of the OpenMC neutron and photon transport Monte Carlo code for Intel GPUs, specifically the Intel Data Center Max 1100 GPU (codename Ponte Vecchio, PVC), through distributed OpenMP offloading. Building upon prior work by Tramm J.R., et al. (2022), which laid the groundwork for GPU adaptation, our study meticulously extends the OpenMC code's capabilities to Intel GPUs. We present a comprehensive benchmarking and scaling analysis, comparing performance on Intel MAX GPUs to state-of-the-art CPU execution (Intel Xeon Platinum 8480+ Processor, codename 4th generation Sapphire Rapids). The results demonstrate a remarkable acceleration factor compared to CPU execution, showcasing the GPU-adapted code's superiority over its CPU counterpart as computational load increases.
- On the inner workings of monte carlo codes. Molecular Simulation, 39(14-15):1253–1292, 2013.
- Sampling algorithms, from survey sampling to monte carlo methods: Tutorial and literature review. arXiv preprint arXiv:2011.00901, 2020.
- Alex F Bielajew. History of monte carlo. In Monte Carlo techniques in radiation therapy, pages 3–15. CRC Press, 2021.
- Andrew L Fielding. Monte-carlo techniques for radiotherapy applications i: introduction and overview of the different monte-carlo codes. Journal of Radiotherapy in Practice, 22:e80, 2023.
- The monte carlo performance benchmark test-aims, specifications and first results. In International Conference on Mathematics and Computational Methods Applied to, volume 2, page 15, 2011.
- Paul Kollath Romano. Parallel algorithms for Monte Carlo particle transport simulation on exascale computing architectures. PhD thesis, Massachusetts Institute of Technology, 2013.
- Direct simulation monte carlo on petaflop supercomputers and beyond. Physics of Fluids, 31(8), 2019.
- Ryan Bleile. The state of monte carlo neutron transport: The role of gpus and portable performance abstractions.
- Exascale multiphysics nuclear reactor simulations for advanced designs. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, pages 1–11, 2023.
- David Schneider. The exascale era is upon us: The frontier supercomputer may be the first to reach 1,000,000,000,000,000,000 operations per second. IEEE spectrum, 59(1):34–35, 2022.
- Implementation, capabilities, and benchmarking of shift, a massively parallel monte carlo radiation transport code. Journal of Computational Physics, 308:239–272, 2016.
- Domain decomposition in the gpu-accelerated shift monte carlo code. Annals of Nuclear Energy, 166:108687, 2022.
- Verification and performance impact of the new parallel mcnp6. 3 particle track output capability for subcritical multiplication simulations. Technical report, Los Alamos National Laboratory (LANL), Los Alamos, NM (United States), 2023.
- Development and verification of geant4-based parallel computing monte carlo simulations for nuclear logging applications. Annals of Nuclear Energy, 172:109079, 2022.
- Jiankai Yu. Nuclear physics probability code: Openmc. In Nuclear Power Plant Design and Analysis Codes, pages 123–138. Elsevier, 2021.
- The openmc monte carlo particle transport code. Annals of Nuclear Energy, 51:274–281, 2013.
- Openmc: A state-of-the-art monte carlo code for research and development. Annals of Nuclear Energy, 82:90–97, 2015.
- Multi-core performance studies of a monte carlo neutron transport code. The International journal of high performance computing applications, 28(1):87–96, 2014.
- OpenMC: Running in Parallel. https://docs.openmc.org/en/v0.13.0/usersguide/parallel.html. [Online].
- Openneomc: A framework for design optimization in particle transport simulations based on openmc and neorl. Annals of Nuclear Energy, 180:109450, 2023.
- Neorl: Neuroevolution optimization with reinforcement learning—applications to carbon-free energy systems. Nuclear Engineering and Design, page 112423, 2023.
- Optimizing the fixed number detector placement for the nuclear reactor core using reinforcement learning. Nuclear Science and Engineering, pages 1–23, 2024.
- Programming Your GPU with OpenMP: Performance Portability for GPUs. MIT Press, 2023.
- Portability and scalability of openmp offloading on state-of-the-art accelerators. arXiv preprint arXiv:2304.04276, 2023.
- Latency and bandwidth microbenchmarks of us department of energy systems in the june 2023 top 500 list. In Proceedings of the SC’23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis, pages 1298–1305, 2023.
- Toward portable gpu acceleration of the openmc monte carlo particle transport code. In International Conference on Physics of Reactors (PHYSOR 2022). Pittsburgh, USA, 2022.
- Intel. Intel Developer Cloud. https://www.intel.com/content/www/us/en/developer/tools/devcloud/overview.html, 2023. [Online].
- Intel. Intel data center gpu max series, 2023.
- Intel. Intel 4th Gen Xeon Scalable Processors. https://www.intel.com/content/www/us/en/newsroom/news/4th-gen-xeon-scalable-processors-max-series-cpus-gpus.html#gs.ti3gm6, 2023. [Online].
- Optimization of neutron tracking algorithms for gpu-based continuous energy monte carlo calculation. Annals of Nuclear Energy, 162:108508, 2021.
- Intel. Intel Data Center GPU Max 1100. https://ark.intel.com/content/www/us/en/ark/products/232876/intel-data-center-gpu-max-1100.html, 2023. [Online].