Towards Exascale Computing for Astrophysical Simulation Leveraging the Leonardo EuroHPC System
Nitin Shukla, Alessandro Romeo, Caterina Caravita, Michael Redenti, Radim Vavrik, Lubomir Riha, Andrea Mignone, Marco Rossazza, Stefano Truzzi, Luca Tornatore, Antonio Ragagnin, Tiago Castro, Geray S. Karademir, Klaus Dolag, Pranab J. Deka, Fabio Bacchini, Rostislav-Paul Wilhelm, Daniele Gregori, Elisabetta Boella
TL;DR
The strategy and initial results on the Leonardo system at CINECA for three flagship codes, namely gPLUTO, OpenGadget3 and iPIC3D, are presented, using profiling tools to analyze performance on single and multiple nodes.
Abstract
Developing and redesigning astrophysical, cosmological, and space plasma numerical codes for existing and next-generation accelerators is critical for enabling large-scale simulations. To address these challenges, the SPACE Center of Excellence (SPACE-CoE) fosters collaboration between scientists, code developers, and high-performance computing experts to optimize applications for the exascale era. This paper presents our strategy and initial results on the Leonardo system at CINECA for three flagship codes, namely gPLUTO, OpenGadget3 and iPIC3D, using profiling tools to analyze performance on single and multiple nodes. Preliminary tests show all three codes scale efficiently, reaching 80% scalability up to 1,024 GPUs.