Implementation and evaluation of an AMR framework for FDM applications

Masaharu Matsumoto, Futoshi Mori, Satoshi Ohshima, Hideyuki Jitsumoto, Takahiro Katagiri, Kengo Nakajima

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

In order to execute various finite-difference method applications on large-scale parallel computers with a reasonable cost of computer resources, a framework using an adaptive mesh refinement (AMR) technique has been developed. AMR can realize high-resolution simulations while saving computer resources by generating and removing hierarchical grids dynamically. In the AMR framework, a dynamic domain decomposition (DDD) technique, as a dynamic load balancing method, is also implemented to correct the computational load imbalance between each process associated with parallelization. By performing a 3D AMR test simulation, it is confirmed that dynamic load balancing can be achieved and execution time can be reduced by introducing the DDD technique.

Original languageEnglish
Pages (from-to)936-946
Number of pages11
JournalProcedia Computer Science
Volume29
DOIs
Publication statusPublished - Jan 1 2014
Event14th Annual International Conference on Computational Science, ICCS 2014 - Cairns, QLD, Australia
Duration: Jun 10 2014Jun 12 2014

Fingerprint

Frequency division multiplexing
Dynamic loads
Resource allocation
Decomposition
Finite difference method
Costs

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Implementation and evaluation of an AMR framework for FDM applications. / Matsumoto, Masaharu; Mori, Futoshi; Ohshima, Satoshi; Jitsumoto, Hideyuki; Katagiri, Takahiro; Nakajima, Kengo.

In: Procedia Computer Science, Vol. 29, 01.01.2014, p. 936-946.

Research output: Contribution to journalConference article

Matsumoto, Masaharu ; Mori, Futoshi ; Ohshima, Satoshi ; Jitsumoto, Hideyuki ; Katagiri, Takahiro ; Nakajima, Kengo. / Implementation and evaluation of an AMR framework for FDM applications. In: Procedia Computer Science. 2014 ; Vol. 29. pp. 936-946.
@article{f04085a9d8dd44b7a89bd3922a26eac6,
title = "Implementation and evaluation of an AMR framework for FDM applications",
abstract = "In order to execute various finite-difference method applications on large-scale parallel computers with a reasonable cost of computer resources, a framework using an adaptive mesh refinement (AMR) technique has been developed. AMR can realize high-resolution simulations while saving computer resources by generating and removing hierarchical grids dynamically. In the AMR framework, a dynamic domain decomposition (DDD) technique, as a dynamic load balancing method, is also implemented to correct the computational load imbalance between each process associated with parallelization. By performing a 3D AMR test simulation, it is confirmed that dynamic load balancing can be achieved and execution time can be reduced by introducing the DDD technique.",
author = "Masaharu Matsumoto and Futoshi Mori and Satoshi Ohshima and Hideyuki Jitsumoto and Takahiro Katagiri and Kengo Nakajima",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.procs.2014.05.084",
language = "English",
volume = "29",
pages = "936--946",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Implementation and evaluation of an AMR framework for FDM applications

AU - Matsumoto, Masaharu

AU - Mori, Futoshi

AU - Ohshima, Satoshi

AU - Jitsumoto, Hideyuki

AU - Katagiri, Takahiro

AU - Nakajima, Kengo

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In order to execute various finite-difference method applications on large-scale parallel computers with a reasonable cost of computer resources, a framework using an adaptive mesh refinement (AMR) technique has been developed. AMR can realize high-resolution simulations while saving computer resources by generating and removing hierarchical grids dynamically. In the AMR framework, a dynamic domain decomposition (DDD) technique, as a dynamic load balancing method, is also implemented to correct the computational load imbalance between each process associated with parallelization. By performing a 3D AMR test simulation, it is confirmed that dynamic load balancing can be achieved and execution time can be reduced by introducing the DDD technique.

AB - In order to execute various finite-difference method applications on large-scale parallel computers with a reasonable cost of computer resources, a framework using an adaptive mesh refinement (AMR) technique has been developed. AMR can realize high-resolution simulations while saving computer resources by generating and removing hierarchical grids dynamically. In the AMR framework, a dynamic domain decomposition (DDD) technique, as a dynamic load balancing method, is also implemented to correct the computational load imbalance between each process associated with parallelization. By performing a 3D AMR test simulation, it is confirmed that dynamic load balancing can be achieved and execution time can be reduced by introducing the DDD technique.

UR - http://www.scopus.com/inward/record.url?scp=84902820908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902820908&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2014.05.084

DO - 10.1016/j.procs.2014.05.084

M3 - Conference article

VL - 29

SP - 936

EP - 946

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -