Hierarchical N-body simulations with auto-tuning for heterogeneous systems

TL;DR

This paper discusses the development of a flexible, auto-tuned N-body simulation framework that leverages hybrid treecodes and FMMs on heterogeneous architectures to enable black-box software solutions.

Contribution

It introduces a novel approach combining hybrid treecodes and FMMs with auto-tuning for heterogeneous systems, advancing the development of versatile black-box N-body simulation libraries.

Findings

Enhanced flexibility of N-body methods through hybridization and auto-tuning

Successful integration of treecodes and FMMs on heterogeneous architectures

Progress towards a black-box software library for fast N-body simulations

Abstract

With the current hybridization of treecodes and FMMs, combined with auto-tuning capabilities on heterogeneous architectures, the flexibility of fast N-body methods has been greatly enhanced. These features are a requirement to developing a black-box software library for fast N-body algorithms on heterogeneous systems, which is our immediate goal.