Skip to Main content Skip to Navigation

Task-based FMM for multicore 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 hardware. In this paper, we propose a new approach that achieves high performance across architectures. Our method consists of expressing the Fast Multipole Method (FMM) algorithm as a task flow and employing a state-of-the-art runtime system, StarPU, to process the tasks on the different computing units. We carefully design the task flow, the mathematical operators, their implementations as well as scheduling schemes. Potentials and forces on 200 million particles are computed in 48.7 seconds on a homogeneous 160 cores SGI Altix UV 100 and good scalability is shown.
Complete list of metadata

Cited literature [14 references]  Display  Hide  Download
Contributor : Olivier Coulaud Connect in order to contact the contributor
Submitted on : Wednesday, April 3, 2013 - 2:23:27 PM
Last modification on : Thursday, January 20, 2022 - 4:15:53 PM
Long-term archiving on: : Sunday, April 2, 2017 - 11:40:35 PM


Files produced by the author(s)


  • HAL Id : hal-00807368, version 1


Emmanuel Agullo, Bérenger Bramas, Olivier Coulaud, Eric Darve, Matthias Messner, et al.. Task-based FMM for multicore architectures. [Research Report] RR-8277, INRIA. 2013, pp.33. ⟨hal-00807368⟩



Les métriques sont temporairement indisponibles