Porting WarpX to GPU-accelerated platforms

• DOI: 10.1016/j.parco.2021.102833
• Publication Date: 2021-09-14T01:31:20Z
• Authors: A. Myers; A. Almgren; L.D. Amorim; J. Bell; L. Fedeli; L. Ge; K. Gott; D.P. Grote; M. Hogan; A. Huebl; R. Jambunathan; R. Lehe; C. Ng; M. Rowan; O. Shapoval; M. Thévenet; J.-L. Vay; H. Vincenti; E. Yang; N. Zaïm; W. Zhang; Y. Zhao; E. Zoni
• Keywords: Porting; Benchmark (surveying); Summit; x86; Code (set theory); Stencil

WarpX is a general purpose electromagnetic particle-in-cell code that was originally designed to run on many-core CPU architectures. We describe the strategy, based AMReX library, followed allow use GPU-accelerated nodes OLCF’s Summit supercomputer, strategy we believe will extend upcoming machines Frontier and Aurora. summarize challenges encountered, lessons learned, give current performance results series of relevant benchmark problems.