First distributed performance-portable NUFFT scales to 1024 GPUs on heterogeneous systems and supports large particle-in-Fourier plasma simulations.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.CE 2years
2026 2verdicts
UNVERDICTED 2roles
method 1polarities
use method 1representative citing papers
FFT-based PIC is fastest in wall time but limited in use, while PCG, FEM, and PIF alternatives cost more yet scale comparably across AMD and Nvidia GPUs.
citing papers explorer
-
A Performance-Portable, Massively Parallel Distributed Nonuniform FFT
First distributed performance-portable NUFFT scales to 1024 GPUs on heterogeneous systems and supports large particle-in-Fourier plasma simulations.
-
A Comparison of Massively Parallel Performance Portable Particle-in-Cell schemes for electrostatic kinetic plasma simulations
FFT-based PIC is fastest in wall time but limited in use, while PCG, FEM, and PIF alternatives cost more yet scale comparably across AMD and Nvidia GPUs.