The PLUTO Code on GPUs: Offloading Lagrangian Particle Methods

Abstract

The Lagrangian Particles (LP) module of the PLUTO code offers a powerful simulation tool to predict the non-thermal emission produced by shock accelerated particles in large-scale relativistic magnetized astrophysics flows. The LPs represent ensembles of relativistic particles with a given energy distribution which is updated by solving the relativistic cosmic ray transport equation. The approach consistently includes the effects of adiabatic expansion, synchrotron and inverse Compton emission. The large scale nature of such systems creates boundless computational demand which can only be satisfied by targeting modern computing hardware such as Graphic Processing Units (GPUs). In this work we presents the GPU-compatible C++ re-design of the LP module, that, by means of the programming model OpenACC and the Message Passing Interface library, is capable of targeting both single commercial GPUs as well as multi-node (pre-)exascale computing facilities. The code has been benchmarked up to 28672 parallel CPUs cores and 1024 parallel GPUs demonstrating $\sim (80-90)\%$ weak scaling parallel efficiency and good strong scaling capabilities. Our results demonstrated a speedup of $6$ times when solving that same benchmark test with 128 full GPU nodes (4GPUs per node) against the same amount of full high-end CPU nodes (112 cores per node). Furthermore, we conducted a code verification by comparing its prediction to corresponding analytical solutions for two test cases. We note that this work is part of broader project that aims at developing gPLUTO, the novel and revised GPU-ready implementation of its legacy.

Published in Astronomy and Computing. Special issue: Advancing Cosmology and Astrophysics through High-Performance Computing and Machine Learning

References in corpus (14)

• PLUTO: a Numerical Code for Computational Astrophysics
• The spectrum of particles accelerated in relativistic, collisionless shocks
• Magnetic reconnection in the era of exascale computing and multiscale experiments
• Spectral evolution of superluminal components in parsec-scale jets
• Shear acceleration in relativistic astrophysical jets
• A Particle Module for the PLUTO code: II - Hybrid Framework for Modeling Non-thermal emission from Relativistic Magnetized flows
• A Particle Module for the PLUTO Code: I - an implementation of the MHD-PIC equations
• Simulating the dynamics and synchrotron emission from relativistic jets II. Evolution of non-thermal electrons
• Two-way coupling of magnetohydrodynamic simulations with embedded particle-in-cell simulations
• Systematic construction of upwind constrained transport schemes for MHD
• Rapid cosmic-ray acceleration at perpendicular shocks in supernova remnants
• Modelling X-shaped Radio Galaxies: Dynamical and Emission Signatures from the Back-flow model
• The polarization of the synchrotron radiation from a recollimated jet: application to high-energy BL Lacs
• A numerical study of the interplay between Fermi acceleration mechanisms in radio lobes of FR-II radio galaxies