Performance-Portable GPU Acceleration of the EFIT Tokamak Plasma Equilibrium Reconstruction Code

Conference · Sat Nov 11 23:00:00 EST 2023 · Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis

DOI:https://doi.org/10.1145/3624062.3624607· OSTI ID:2477210

Antepara, Oscar ^[1]; Williams, Samuel ^[2]; Kruger, Scott ^[3]; Bechtel, Torrin ^[4]; McClenaghan, Joseph ^[4]; Lao, Lang ^[4]

---

1. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); General Atomics
2. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
3. Tech-X Corporation, Boulder, CO (United States)
4. General Atomics, San Diego, CA (United States)

This paper presents the steps followed to GPU-offload parts of the core solver of EFIT-AI, an equilibrium reconstruction code suitable for tokamak experiments and burning plasmas. For this work, we will focus on the fitting procedure that consists of a Grad–Shafranov (GS) equation inverse solver that calculates equilibrium reconstructions on a grid. We will show profiling results of the original code (CPU-baseline), as well as the directives used to GPU-offload the most time-consuming function, initially to compare OpenACC and OpenMP on NVIDIA and AMD GPUs and later on to assess OpenMP performance portability on NVIDIA, AMD and Intel GPUs. We will make a performance comparison for different spatial grid sizes and show the speedup achieved on NVIDIA A100 (Perlmutter-NERSC), AMD MI250X (Frontier-OLCF) and Intel PVC GPUs (Sunspot-ALCF). Finally, we will draw some conclusions and recommendations to achieve high-performance portability for an equilibrium reconstruction code on the new HPC architectures

• Research Organization: General Atomics, San Diego, CA (United States)
• Sponsoring Organization: USDOE Office of Science (SC), Fusion Energy Sciences (FES); USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF)
• DOE Contract Number: FC02-04ER54698; SC0021203; AC02-05CH11231; AC05-00OR22725; AC02-06CH11357
• OSTI ID: 2477210
• Journal Information: Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis, Journal Name: Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis
• Country of Publication: United States
• Language: English

Related Subjects

GPU
GPU offloading directives
OpenACC
OpenMP
performance portability