Semi-Lagrangian 4d, 5d, and 6d kinetic plasma simulation on large-scale GPU-equipped supercomputers

Running kinetic plasma physics simulations using grid-based solvers is very demanding both in terms of memory as well as computational cost. This is primarily due to the up to six-dimensional phase space and the associated unfavorable scaling of the computational cost as a function of grid spacing (...

Full description

Saved in:
Bibliographic Details
Published in:The international journal of high performance computing applications 2023-03, Vol.37 (2), p.180-196
Main Authors: Einkemmer, Lukas, Moriggl, Alexander
Format: Article
Language:English
Subjects:
Computational efficiency
Computing costs
Graphics processing units
High performance computing
Plasma physics
Poisson equation
Simulation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Running kinetic plasma physics simulations using grid-based solvers is very demanding both in terms of memory as well as computational cost. This is primarily due to the up to six-dimensional phase space and the associated unfavorable scaling of the computational cost as a function of grid spacing (often termed the curse of dimensionality). In this article, we present 4d, 5d, and 6d simulations of the Vlasov–Poisson equation with a split-step semi-Lagrangian discontinuous Galerkin scheme on graphic processing units (GPUs). The local communication pattern of this method allows an efficient implementation on large-scale GPU-based systems and emphasizes the importance of considering algorithmic and high-performance computing aspects in unison. We demonstrate a single node performance above 2 TB/s effective memory bandwidth (on a node with four A100 GPUs) and show excellent scaling (parallel efficiency between 30% and 67%) for up to 1536 A100 GPUs on JUWELS Booster. Graphical Abstract
ISSN:1094-3420
1741-2846
DOI:10.1177/10943420221137599