A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries

M. J. Aftosmis, Marsha Berger, G. Adomavicius

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Preliminary verification and validation of an efficient Euler solver for adaptively refined Cartesian meshes with embedded boundaries is presented. The parallel, multilevel method makes use of a new on-the-fly parallel domain decomposition strategy based upon the use of space-filling curves, and automatically generates a sequence of coarse meshes for processing by the multigrid smoother. The coarse mesh generation algorithm produces grids which completely cover the computational domain at every level in the mesh hierarchy. A series of examples on realistically complex three-dimensional configurations demonstrate that this new coarsening algorithm reliably achieves mesh coarsening ratios in excess of 7 on adaptively refined meshes. Numerical investigations of the scheme's local truncation error demonstrate an achieved order of accuracy between 1.82 and 1.88. Convergence results for the multigrid scheme are presented for both subsonic and transonic test cases and demonstrate W-cycle multigrid convergence rates between 0.84 and 0.94. Preliminary parallel scalability tests on both simple wing and complex complete aircraft geometries show a computational speedup of 52 using 64 processors with the run-time mesh partitioner.

Original languageEnglish (US)
Title of host publication38th Aerospace Sciences Meeting and Exhibit
StatePublished - 2000
Event38th Aerospace Sciences Meeting and Exhibit 2000 - Reno, NV, United States
Duration: Jan 10 2000Jan 13 2000

Other

Other38th Aerospace Sciences Meeting and Exhibit 2000
CountryUnited States
CityReno, NV
Period1/10/001/13/00

Fingerprint

Coarsening
mesh
grids
Mesh generation
Scalability
aircraft
Aircraft
decomposition
Decomposition
geometry
Geometry
Processing
truncation errors
transonic flow
wings
hierarchies
central processing units
method
test
cycles

ASJC Scopus subject areas

  • Space and Planetary Science
  • Aerospace Engineering

Cite this

Aftosmis, M. J., Berger, M., & Adomavicius, G. (2000). A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries. In 38th Aerospace Sciences Meeting and Exhibit

A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries. / Aftosmis, M. J.; Berger, Marsha; Adomavicius, G.

38th Aerospace Sciences Meeting and Exhibit. 2000.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Aftosmis, MJ, Berger, M & Adomavicius, G 2000, A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries. in 38th Aerospace Sciences Meeting and Exhibit. 38th Aerospace Sciences Meeting and Exhibit 2000, Reno, NV, United States, 1/10/00.
Aftosmis MJ, Berger M, Adomavicius G. A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries. In 38th Aerospace Sciences Meeting and Exhibit. 2000
Aftosmis, M. J. ; Berger, Marsha ; Adomavicius, G. / A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries. 38th Aerospace Sciences Meeting and Exhibit. 2000.
@inproceedings{2badbda3e4734baba96eea3da85ee50f,
title = "A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries",
abstract = "Preliminary verification and validation of an efficient Euler solver for adaptively refined Cartesian meshes with embedded boundaries is presented. The parallel, multilevel method makes use of a new on-the-fly parallel domain decomposition strategy based upon the use of space-filling curves, and automatically generates a sequence of coarse meshes for processing by the multigrid smoother. The coarse mesh generation algorithm produces grids which completely cover the computational domain at every level in the mesh hierarchy. A series of examples on realistically complex three-dimensional configurations demonstrate that this new coarsening algorithm reliably achieves mesh coarsening ratios in excess of 7 on adaptively refined meshes. Numerical investigations of the scheme's local truncation error demonstrate an achieved order of accuracy between 1.82 and 1.88. Convergence results for the multigrid scheme are presented for both subsonic and transonic test cases and demonstrate W-cycle multigrid convergence rates between 0.84 and 0.94. Preliminary parallel scalability tests on both simple wing and complex complete aircraft geometries show a computational speedup of 52 using 64 processors with the run-time mesh partitioner.",
author = "Aftosmis, {M. J.} and Marsha Berger and G. Adomavicius",
year = "2000",
language = "English (US)",
booktitle = "38th Aerospace Sciences Meeting and Exhibit",

}

TY - GEN

T1 - A parallel multilevel method for adaptively refined Cartesian grids with embedded boundaries

AU - Aftosmis, M. J.

AU - Berger, Marsha

AU - Adomavicius, G.

PY - 2000

Y1 - 2000

N2 - Preliminary verification and validation of an efficient Euler solver for adaptively refined Cartesian meshes with embedded boundaries is presented. The parallel, multilevel method makes use of a new on-the-fly parallel domain decomposition strategy based upon the use of space-filling curves, and automatically generates a sequence of coarse meshes for processing by the multigrid smoother. The coarse mesh generation algorithm produces grids which completely cover the computational domain at every level in the mesh hierarchy. A series of examples on realistically complex three-dimensional configurations demonstrate that this new coarsening algorithm reliably achieves mesh coarsening ratios in excess of 7 on adaptively refined meshes. Numerical investigations of the scheme's local truncation error demonstrate an achieved order of accuracy between 1.82 and 1.88. Convergence results for the multigrid scheme are presented for both subsonic and transonic test cases and demonstrate W-cycle multigrid convergence rates between 0.84 and 0.94. Preliminary parallel scalability tests on both simple wing and complex complete aircraft geometries show a computational speedup of 52 using 64 processors with the run-time mesh partitioner.

AB - Preliminary verification and validation of an efficient Euler solver for adaptively refined Cartesian meshes with embedded boundaries is presented. The parallel, multilevel method makes use of a new on-the-fly parallel domain decomposition strategy based upon the use of space-filling curves, and automatically generates a sequence of coarse meshes for processing by the multigrid smoother. The coarse mesh generation algorithm produces grids which completely cover the computational domain at every level in the mesh hierarchy. A series of examples on realistically complex three-dimensional configurations demonstrate that this new coarsening algorithm reliably achieves mesh coarsening ratios in excess of 7 on adaptively refined meshes. Numerical investigations of the scheme's local truncation error demonstrate an achieved order of accuracy between 1.82 and 1.88. Convergence results for the multigrid scheme are presented for both subsonic and transonic test cases and demonstrate W-cycle multigrid convergence rates between 0.84 and 0.94. Preliminary parallel scalability tests on both simple wing and complex complete aircraft geometries show a computational speedup of 52 using 64 processors with the run-time mesh partitioner.

UR - http://www.scopus.com/inward/record.url?scp=2942702401&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942702401&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:2942702401

BT - 38th Aerospace Sciences Meeting and Exhibit

ER -