High Performance P3M N-body code: CUBEP3M

TL;DR

CUBEP3M is a high-performance, scalable cosmological N-body simulation code with utilities for large-scale structure studies, optimized for speed and memory efficiency, suitable for diverse astrophysical applications.

Contribution

The paper introduces CUBEP3M, a publicly available N-body code with novel utilities, enhanced scalability, and optimized memory usage for large cosmological simulations.

Findings

Achieves near-ideal weak scaling on 27,000 cores

Memory footprint as low as 37 bytes per particle in lean mode

Effective for large-scale cosmological applications

Abstract

This paper presents CUBEP3M, a publicly-available high performance cosmological N-body code and describes many utilities and extensions that have been added to the standard package. These include a memory-light runtime SO halo finder, a non-Gaussian initial conditions generator, and a system of unique particle identification. CUBEP3M is fast, its accuracy is tuneable to optimize speed or memory, and has been run on more than 27,000 cores, achieving within a factor of two of ideal weak scaling even at this problem size. The code can be run in an extra-lean mode where the peak memory imprint for large runs is as low as 37 bytes per particles, which is almost two times leaner than other widely used N-body codes. However, load imbalances can increase this requirement by a factor of two, such that fast configurations with all the utilities enabled and load imbalances factored in require between 70 and 120 bytes per particles. CUBEP3M is well designed to study large scales cosmological systems, where imbalances are not too large and adaptive time-stepping not essential. It has already been used for a broad number of science applications that require either large samples of non-linear realizations or very large dark matter N-body simulations, including cosmological reionization, halo formation, baryonic acoustic oscillations, weak lensing or non-Gaussian statistics. We discuss the structure, the accuracy, known systematic effects and the scaling performance of the code and its utilities, when applicable.