Parallelization of moving particle semi-implicit method

Guangtao Duan, Bin Chen

Research output: Contribution to journalArticle

Abstract

The Moving Particle Semi-implicit Method (MPS) has been proven useful in simulation of free-surface flows. One of its main drawbacks is the high computational load. The solution for the Pressure Poisson Equation (PPE) is the critical issue in parallelizing the MPS algorithm because it is the most time-consuming part and the original ICCG solver is difficult to parallelize. The Symmetric Lanczos Algorithm (SLA), which is easy to parallelize and as effective as ICCG, is introduced to solve PPE. After decomposing computational domain into subdomains which are one by one assigned to different nodes of a cluster, the MPI model is used to exchange the information of particles between the adjacent subdomains and the OpenMP model is used to accelerate the simulation in each node. It is shown that MPS can be accelerated by 37.5 times on a cluster with 100 CPU cores by the hybrid OpenMP/MPI model.

Original languageEnglish
Pages (from-to)2077-2080
Number of pages4
JournalKung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics
Volume34
Issue number11
Publication statusPublished - 2013 Nov 1
Externally publishedYes

Fingerprint

Poisson equation
Program processors
simulation

Keywords

  • MPS
  • OpenMP/MPI
  • Pressure poisson equation
  • Symmetric Lanczos algorithm

ASJC Scopus subject areas

  • Materials Science(all)
  • Condensed Matter Physics
  • Mechanical Engineering

Cite this

Parallelization of moving particle semi-implicit method. / Duan, Guangtao; Chen, Bin.

In: Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics, Vol. 34, No. 11, 01.11.2013, p. 2077-2080.

Research output: Contribution to journalArticle

@article{54392fb7b7cc483ba367440f37a2dfbe,
title = "Parallelization of moving particle semi-implicit method",
abstract = "The Moving Particle Semi-implicit Method (MPS) has been proven useful in simulation of free-surface flows. One of its main drawbacks is the high computational load. The solution for the Pressure Poisson Equation (PPE) is the critical issue in parallelizing the MPS algorithm because it is the most time-consuming part and the original ICCG solver is difficult to parallelize. The Symmetric Lanczos Algorithm (SLA), which is easy to parallelize and as effective as ICCG, is introduced to solve PPE. After decomposing computational domain into subdomains which are one by one assigned to different nodes of a cluster, the MPI model is used to exchange the information of particles between the adjacent subdomains and the OpenMP model is used to accelerate the simulation in each node. It is shown that MPS can be accelerated by 37.5 times on a cluster with 100 CPU cores by the hybrid OpenMP/MPI model.",
keywords = "MPS, OpenMP/MPI, Pressure poisson equation, Symmetric Lanczos algorithm",
author = "Guangtao Duan and Bin Chen",
year = "2013",
month = "11",
day = "1",
language = "English",
volume = "34",
pages = "2077--2080",
journal = "Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics",
issn = "0253-231X",
publisher = "Science Press",
number = "11",

}

TY - JOUR

T1 - Parallelization of moving particle semi-implicit method

AU - Duan, Guangtao

AU - Chen, Bin

PY - 2013/11/1

Y1 - 2013/11/1

N2 - The Moving Particle Semi-implicit Method (MPS) has been proven useful in simulation of free-surface flows. One of its main drawbacks is the high computational load. The solution for the Pressure Poisson Equation (PPE) is the critical issue in parallelizing the MPS algorithm because it is the most time-consuming part and the original ICCG solver is difficult to parallelize. The Symmetric Lanczos Algorithm (SLA), which is easy to parallelize and as effective as ICCG, is introduced to solve PPE. After decomposing computational domain into subdomains which are one by one assigned to different nodes of a cluster, the MPI model is used to exchange the information of particles between the adjacent subdomains and the OpenMP model is used to accelerate the simulation in each node. It is shown that MPS can be accelerated by 37.5 times on a cluster with 100 CPU cores by the hybrid OpenMP/MPI model.

AB - The Moving Particle Semi-implicit Method (MPS) has been proven useful in simulation of free-surface flows. One of its main drawbacks is the high computational load. The solution for the Pressure Poisson Equation (PPE) is the critical issue in parallelizing the MPS algorithm because it is the most time-consuming part and the original ICCG solver is difficult to parallelize. The Symmetric Lanczos Algorithm (SLA), which is easy to parallelize and as effective as ICCG, is introduced to solve PPE. After decomposing computational domain into subdomains which are one by one assigned to different nodes of a cluster, the MPI model is used to exchange the information of particles between the adjacent subdomains and the OpenMP model is used to accelerate the simulation in each node. It is shown that MPS can be accelerated by 37.5 times on a cluster with 100 CPU cores by the hybrid OpenMP/MPI model.

KW - MPS

KW - OpenMP/MPI

KW - Pressure poisson equation

KW - Symmetric Lanczos algorithm

UR - http://www.scopus.com/inward/record.url?scp=84889248734&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84889248734&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84889248734

VL - 34

SP - 2077

EP - 2080

JO - Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics

JF - Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics

SN - 0253-231X

IS - 11

ER -