Speaker
Description
Particle–mesh methods such as Particle-in-Cell (PIC) remain central to plasma, beam, and astrophysical simulation. We present the current state of the IPPL (Independent Parallel Particle Layer) library, which provides performance portable and dimension independent building blocks for scientific simulations requiring particle-mesh methods. IPPL makes use of Kokkos, HeFFTe, and MPI (Message Passing Interface) to deliver a portable, massively parallel toolkit supporting simulations in one to six dimensions, mixed precision, and asynchronous execution in different execution spaces (e.g. CPUs and GPUs).
IPPL is very well suited for research on novel numerical solvers, and as showcases we elaborate our recent work on a novel spectrally accurate free-space Poisson solver [1] and IPPL-based mini-apps for kinetic plasma simulations [2]. We achieve high particle throughput on GPUs while maintaining scaling across multiple nodes. The mini-apps also help us to identify the dominant costs in particle migration, halo exchange, and field solvers.
[1] S Mayani et al., A massively parallel performance portable free-space spectral Poisson solver,
ACM Transactions on Mathematical Software (51)3, 2025, https://dl.acm.org/doi/10.1145/3748815
[2] S Muralikrishnan et al., Scaling and performance portability of the particle-in-cell scheme for plasma physics applications through mini-apps targeting exascale architectures,
SIAM Parallel Processing, 2024,\ https://doi.org/10.1137/1.9781611977967.3 and \ https://arxiv.org/abs/2205.11052
