Implementation and Performance of Barnes-Hut N-body algorithm on Extreme-scale Heterogeneous Many-core Architectures

TL;DR

This paper presents the implementation and performance analysis of the Barnes-Hut N-body algorithm on extreme-scale heterogeneous many-core architectures, demonstrating high efficiency on systems with billions of cores.

Contribution

The paper introduces new algorithms for efficient Barnes-Hut N-body simulations on systems with low memory bandwidth and extreme core counts, integrated into the FDPS framework.

Findings

Achieved 47.9 PF on TaihuLight with 10 million cores

Achieved 10.6 PF on Gyoukou with 16 million cores

Achieved 1.01 PF on Shoubu System B with 10 million cores

Abstract

In this paper, we report the implementation and measured performance of our extreme-scale global simulation code on Sunway TaihuLight and two PEZY-SC2 systems: Shoubu System B and Gyoukou. The numerical algorithm is the parallel Barnes-Hut tree algorithm, which has been used in many large-scale astrophysical particle-based simulations. Our implementation is based on our FDPS framework. However, the extremely large numbers of cores of the systems used (10M on TaihuLight and 16M on Gyoukou) and their relatively poor memory and network bandwidth pose new challenges. We describe the new algorithms introduced to achieve high efficiency on machines with low memory bandwidth. The measured performance is 47.9, 10.6 PF, and 1.01PF on TaihuLight, Gyoukou and Shoubu System B (efficiency 40\%, 23.5\% and 35.5\%). The current code is developed for the simulation of planetary rings, but most of the new algorithms are useful for other simulations, and are now available in the FDPS framework.