Pipelining the Fast Multipole Method over a Runtime System

TL;DR

This paper presents a novel approach to accelerate Fast Multipole Methods by expressing them as task flows and leveraging a runtime system to efficiently utilize heterogeneous hardware, achieving high performance across architectures.

Contribution

It introduces a task flow-based implementation of FMM using StarPU, enabling high-performance execution on diverse hardware without extensive tuning.

Findings

200 million particles computed in 48.7 seconds on SGI Altix UV 100

38 million particles computed in 13.34 seconds on a heterogeneous system

Demonstrates portability and efficiency across different architectures

Abstract

Fast Multipole Methods (FMM) are a fundamental operation for the simulation of many physical problems. The high performance design of such methods usually requires to carefully tune the algorithm for both the targeted physics and the hardware. In this paper, we propose a new approach that achieves high performance across architectures. Our method consists of expressing the FMM algorithm as a task flow and employing a state-of-the-art runtime system, StarPU, in order to process the tasks on the different processing units. We carefully design the task flow, the mathematical operators, their Central Processing Unit (CPU) and Graphics Processing Unit (GPU) implementations, as well as scheduling schemes. We compute potentials and forces of 200 million particles in 48.7 seconds on a homogeneous 160 cores SGI Altix UV 100 and of 38 million particles in 13.34 seconds on a heterogeneous 12 cores Intel Nehalem processor enhanced with 3 Nvidia M2090 Fermi GPUs.