Acceleration of hybrid MPI parallel NBODY6++ for large N-body globular cluster simulations

Long Wang; Rainer Spurzem; Sverre Aarseth; Keigo Nitadori; Peter Berczik; M. B.N. Kouwenhoven; Thorsten Naab

doi:10.1017/S174392131500798X

Acceleration of hybrid MPI parallel NBODY6++ for large N-body globular cluster simulations

Long Wang, Rainer Spurzem, Sverre Aarseth, Keigo Nitadori, Peter Berczik, M. B.N. Kouwenhoven, Thorsten Naab

Research output: Contribution to journal › Article › peer-review

Abstract

Previous research on globular clusters (GCs) dynamics is mostly based on semi-analytic, Fokker-Planck, Monte-Carlo methods and on direct N-body (NB) simulations. These works have great advantages but also limits since GCs are massive and compact and close encounters and binaries play very important roles in their dynamics. The former three methods make approximations and assumptions, while expensive computing time and number of stars limit the latter method. The current largest direct NB simulation has ∼ 500k stars (Heggie 2014). Here, we accelerate the direct NB code NBODY6++ (which extends NBODY6 to supercomputers by using MPI) with new parallel computing technologies (GPU, OpenMP + SSE/AVX). Our aim is to handle large N (up to 10⁶) direct NB simulations to obtain better understanding of the dynamical evolution of GCs.

Original language	English
Pages (from-to)	260-261
Number of pages	2
Journal	Proceedings of the International Astronomical Union
Volume	10
DOIs	https://doi.org/10.1017/S174392131500798X
Publication status	Published - 2016
Externally published	Yes

Keywords

globular clusters: general
methods: numerical

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1017/S174392131500798X

Cite this

@article{c991989f177149039fd22beac4e29ddd,

title = "Acceleration of hybrid MPI parallel NBODY6++ for large N-body globular cluster simulations",

abstract = "Previous research on globular clusters (GCs) dynamics is mostly based on semi-analytic, Fokker-Planck, Monte-Carlo methods and on direct N-body (NB) simulations. These works have great advantages but also limits since GCs are massive and compact and close encounters and binaries play very important roles in their dynamics. The former three methods make approximations and assumptions, while expensive computing time and number of stars limit the latter method. The current largest direct NB simulation has ∼ 500k stars (Heggie 2014). Here, we accelerate the direct NB code NBODY6++ (which extends NBODY6 to supercomputers by using MPI) with new parallel computing technologies (GPU, OpenMP + SSE/AVX). Our aim is to handle large N (up to 106) direct NB simulations to obtain better understanding of the dynamical evolution of GCs.",

keywords = "globular clusters: general, methods: numerical",

author = "Long Wang and Rainer Spurzem and Sverre Aarseth and Keigo Nitadori and Peter Berczik and Kouwenhoven, {M. B.N.} and Thorsten Naab",

note = "Publisher Copyright: Copyright {\textcopyright} 2016 International Astronomical Union.",

year = "2016",

doi = "10.1017/S174392131500798X",

language = "English",

volume = "10",

pages = "260--261",

journal = "Proceedings of the International Astronomical Union",

issn = "1743-9213",

}

TY - JOUR

T1 - Acceleration of hybrid MPI parallel NBODY6++ for large N-body globular cluster simulations

AU - Wang, Long

AU - Spurzem, Rainer

AU - Aarseth, Sverre

AU - Nitadori, Keigo

AU - Berczik, Peter

AU - Kouwenhoven, M. B.N.

AU - Naab, Thorsten

PY - 2016

Y1 - 2016

N2 - Previous research on globular clusters (GCs) dynamics is mostly based on semi-analytic, Fokker-Planck, Monte-Carlo methods and on direct N-body (NB) simulations. These works have great advantages but also limits since GCs are massive and compact and close encounters and binaries play very important roles in their dynamics. The former three methods make approximations and assumptions, while expensive computing time and number of stars limit the latter method. The current largest direct NB simulation has ∼ 500k stars (Heggie 2014). Here, we accelerate the direct NB code NBODY6++ (which extends NBODY6 to supercomputers by using MPI) with new parallel computing technologies (GPU, OpenMP + SSE/AVX). Our aim is to handle large N (up to 106) direct NB simulations to obtain better understanding of the dynamical evolution of GCs.

AB - Previous research on globular clusters (GCs) dynamics is mostly based on semi-analytic, Fokker-Planck, Monte-Carlo methods and on direct N-body (NB) simulations. These works have great advantages but also limits since GCs are massive and compact and close encounters and binaries play very important roles in their dynamics. The former three methods make approximations and assumptions, while expensive computing time and number of stars limit the latter method. The current largest direct NB simulation has ∼ 500k stars (Heggie 2014). Here, we accelerate the direct NB code NBODY6++ (which extends NBODY6 to supercomputers by using MPI) with new parallel computing technologies (GPU, OpenMP + SSE/AVX). Our aim is to handle large N (up to 106) direct NB simulations to obtain better understanding of the dynamical evolution of GCs.

KW - globular clusters: general

KW - methods: numerical

UR - http://www.scopus.com/inward/record.url?scp=84960372201&partnerID=8YFLogxK

U2 - 10.1017/S174392131500798X

DO - 10.1017/S174392131500798X

M3 - Article

AN - SCOPUS:84960372201

SN - 1743-9213

VL - 10

SP - 260

EP - 261

JO - Proceedings of the International Astronomical Union

JF - Proceedings of the International Astronomical Union

ER -

Acceleration of hybrid MPI parallel NBODY6++ for large N-body globular cluster simulations

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Cite this