A multithreaded OpenMP implementation of the LU-SGS method using the multilevel decomposition of the unstructured computational mesh

M. N. Petrov; V. A. Titarev; S. V. Utyuzhnikov; A. V. Chikitkin

doi:10.1134/S0965542517110124

A multithreaded OpenMP implementation of the LU-SGS method using the multilevel decomposition of the unstructured computational mesh

M. N. Petrov^*, V. A. Titarev, S. V. Utyuzhnikov, A. V. Chikitkin

^*Corresponding author for this work

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

Abstract

A new parallel version of the method LU-SGS (Lower-upper symmetric Gauss-Seidel) based on a multilevel decomposition of the unstructured computational mesh is proposed. The advantages of the proposed approach are demonstrated by computing the supersonic flow around the RAM-C geometry. The method is well scalable when a large number of threads are used on the processor Intel Xeon Phi.

Original language	English
Pages (from-to)	1856-1865
Number of pages	10
Journal	Computational Mathematics and Mathematical Physics
Volume	57
Issue number	11
Early online date	28 Nov 2017
DOIs	https://doi.org/10.1134/S0965542517110124
Publication status	Published - Nov 2017

Keywords

computational fluid dynamics
hypersonic flows
Intel Xeon Phi
multithreaded LU-SGS scheme
unstructured mesh

Access to Document

10.1134/S0965542517110124

Cite this

@article{6cd69c65c95040138e81fa67576206f3,

title = "A multithreaded OpenMP implementation of the LU-SGS method using the multilevel decomposition of the unstructured computational mesh",

abstract = "A new parallel version of the method LU-SGS (Lower-upper symmetric Gauss-Seidel) based on a multilevel decomposition of the unstructured computational mesh is proposed. The advantages of the proposed approach are demonstrated by computing the supersonic flow around the RAM-C geometry. The method is well scalable when a large number of threads are used on the processor Intel Xeon Phi.",

keywords = "computational fluid dynamics, hypersonic flows, Intel Xeon Phi, multithreaded LU-SGS scheme, unstructured mesh",

author = "Petrov, {M. N.} and Titarev, {V. A.} and Utyuzhnikov, {S. V.} and Chikitkin, {A. V.}",

year = "2017",

month = nov,

doi = "10.1134/S0965542517110124",

language = "English",

volume = "57",

pages = "1856--1865",

journal = "Computational Mathematics and Mathematical Physics",

issn = "0965-5425",

publisher = "Maik Nauka-Interperiodica Publishing",

number = "11",

}

TY - JOUR

T1 - A multithreaded OpenMP implementation of the LU-SGS method using the multilevel decomposition of the unstructured computational mesh

AU - Petrov, M. N.

AU - Titarev, V. A.

AU - Utyuzhnikov, S. V.

AU - Chikitkin, A. V.

PY - 2017/11

Y1 - 2017/11

N2 - A new parallel version of the method LU-SGS (Lower-upper symmetric Gauss-Seidel) based on a multilevel decomposition of the unstructured computational mesh is proposed. The advantages of the proposed approach are demonstrated by computing the supersonic flow around the RAM-C geometry. The method is well scalable when a large number of threads are used on the processor Intel Xeon Phi.

AB - A new parallel version of the method LU-SGS (Lower-upper symmetric Gauss-Seidel) based on a multilevel decomposition of the unstructured computational mesh is proposed. The advantages of the proposed approach are demonstrated by computing the supersonic flow around the RAM-C geometry. The method is well scalable when a large number of threads are used on the processor Intel Xeon Phi.

KW - computational fluid dynamics

KW - hypersonic flows

KW - Intel Xeon Phi

KW - multithreaded LU-SGS scheme

KW - unstructured mesh

UR - http://www.scopus.com/inward/record.url?scp=85037033081&partnerID=8YFLogxK

U2 - 10.1134/S0965542517110124

DO - 10.1134/S0965542517110124

M3 - Article

AN - SCOPUS:85037033081

SN - 0965-5425

VL - 57

SP - 1856

EP - 1865

JO - Computational Mathematics and Mathematical Physics

JF - Computational Mathematics and Mathematical Physics

IS - 11

ER -

A multithreaded OpenMP implementation of the LU-SGS method using the multilevel decomposition of the unstructured computational mesh

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this