At SC22: ACM Gordon Bell Prize Awarded for Particle-In-Cell Simulations on Frontier, Fugaku, Summit and Perlmutter Supercomputers

Print Friendly, PDF & Email

New York, NY, November 17, 2022 – ACM, the Association for Computing Machinery, named a 16-member team drawn from French, Japanese, and US institutions as recipient of the 2022 ACM Gordon Bell Prize for their project, “Pushing the Frontier in the Design of Laser-Based Electron Accelerators With Groundbreaking Mesh-Refined Particle-In-Cell Simulations on Exascale-Class Supercomputers.”

The members of the team are: Luca Fedeli (CEA), Axel Huebl (Lawrence Berkeley National Laboratory), France Boillod-Cerneaux (CEA), Thomas Clark (CEA), Kevin Gott (Lawrence Berkeley National Laboratory), Conrad Hillairet (Arm), Stephan Jaure (ATOS), Adrien Leblanc (Laboratoire d’Optique Appliquée, ENSTA Paris), Remi Lehe (Lawrence Berkeley National Laboratory) Andrew Myers (Lawrence Berkeley National Laboratory), Christelle Piechurski (GENCI) Mitsuhisa Sato (RIKEN), Neil Zaїm (CEA), Weiqun Zhang (Lawrence Berkeley National Laboratory), Jean-Luc Vay (Lawrence Berkeley National Laboratory), and Henri Vincenti (CEA).

Particle-in-Cell (PIC) simulation is a technique within high-performance computing used to model the motion of charged particles, or plasma. PIC has applications in many areas, including nuclear fusion, accelerators, space physics, and astrophysics. The very recent introduction of exascale-class computers has expanded the horizons of PIC simulations and makes this year’s winning project especially exciting.

According to their abstract, the team presents a first-of-kind mesh-refined (MR) massively parallel PIC code for kinetic plasma simulations optimized on the Frontier, Fugaku, Summit, and Perlmutter supercomputers. The major improvements in their PIC code over existing state-of-the-art approaches include:

  • A three-level parallelization strategy that demonstrated performance portability and scaling on millions of A64FX cores and tens of thousands of AMD and Nvidia GPU’s
  • Fugaku supercomputer

    A groundbreaking mesh refinement capability that provides between 1.5x to 4x savings in computing requirements

  • An efficient load-balancing strategy between multiple mesh-refined levels

Summit supercomputer

The 2022 ACM Gordon Bell Prize-winning team concludes by noting that, “the use of mesh refinement in large-scale electromagnetic PIC simulations is a first and represents a paradigm shift. The successful modeling with savings between 1.5× and 4× with mesh refinement that is reported in this paper is a landmark steppingstone toward a new era in the modelling of laser-plasma interactions.”

The ACM Gordon Bell Prize tracks the progress of parallel computing and rewards innovation in applying high-performance computing to challenges in science, engineering, and large-scale data analytics. The award was presented during the International Conference for High Performance Computing, Networking, Storage and Analysis (SC22), which was held in Dallas, Texas.