Parallel adaptive solvers in compressible petsc-fun3d simulations

S. Bhowmick, D. Kaushik, L. McInnes, B. Norris, P. Raghavan

Research output: Chapter in Book/Report/Conference proceedingChapter

5 Citations (Scopus)

Abstract

The chapter presents a polyalgorithmic technique for adaptively selecting the linear solver method to match the numeric properties of linear systems as they evolve during the course of nonlinear iterations. The approach combines more robust but more costly methods when needed in particularly challenging phases of solution, with cheaper, though less powerful, methods in other phases. The chapter demonstrates that this adaptive, polyalgorithmic approach leads to improvements in overall simulation time, is easily parallelized, and is scalable in the context of this large-scale computational fluid dynamics application. This approach reduced overall execution time by using cheaper, though less powerful, linear solvers for relatively easy linear systems and then switching to more robust but more costly methods for more difficult linear systems. The results demonstrate that adaptive solvers can be implemented easily in a multiprocessor environment and are scalable. The chapter investigates adaptive solvers in problem domains and considers more adaptive approaches, including a polynomial heuristic where the trends of the indicators can be estimated by fitting a function to known data points. The chapter also combines adaptive heuristics with high-performance component infrastructure for performance monitoring and analysis.

Original languageEnglish (US)
Title of host publicationParallel Computational Fluid Dynamics 2005
PublisherElsevier
Pages277-284
Number of pages8
ISBN (Print)9780444522061
DOIs
StatePublished - Dec 1 2006

Fingerprint

Linear systems
Computational fluid dynamics
Polynomials
Monitoring

ASJC Scopus subject areas

  • Chemical Engineering(all)

Cite this

Bhowmick, S., Kaushik, D., McInnes, L., Norris, B., & Raghavan, P. (2006). Parallel adaptive solvers in compressible petsc-fun3d simulations. In Parallel Computational Fluid Dynamics 2005 (pp. 277-284). Elsevier. https://doi.org/10.1016/B978-044452206-1/50033-1

Parallel adaptive solvers in compressible petsc-fun3d simulations. / Bhowmick, S.; Kaushik, D.; McInnes, L.; Norris, B.; Raghavan, P.

Parallel Computational Fluid Dynamics 2005. Elsevier, 2006. p. 277-284.

Research output: Chapter in Book/Report/Conference proceedingChapter

Bhowmick, S, Kaushik, D, McInnes, L, Norris, B & Raghavan, P 2006, Parallel adaptive solvers in compressible petsc-fun3d simulations. in Parallel Computational Fluid Dynamics 2005. Elsevier, pp. 277-284. https://doi.org/10.1016/B978-044452206-1/50033-1
Bhowmick S, Kaushik D, McInnes L, Norris B, Raghavan P. Parallel adaptive solvers in compressible petsc-fun3d simulations. In Parallel Computational Fluid Dynamics 2005. Elsevier. 2006. p. 277-284 https://doi.org/10.1016/B978-044452206-1/50033-1
Bhowmick, S. ; Kaushik, D. ; McInnes, L. ; Norris, B. ; Raghavan, P. / Parallel adaptive solvers in compressible petsc-fun3d simulations. Parallel Computational Fluid Dynamics 2005. Elsevier, 2006. pp. 277-284
@inbook{87bb861ad3cf4b59b6883f37271aa7d4,
title = "Parallel adaptive solvers in compressible petsc-fun3d simulations",
abstract = "The chapter presents a polyalgorithmic technique for adaptively selecting the linear solver method to match the numeric properties of linear systems as they evolve during the course of nonlinear iterations. The approach combines more robust but more costly methods when needed in particularly challenging phases of solution, with cheaper, though less powerful, methods in other phases. The chapter demonstrates that this adaptive, polyalgorithmic approach leads to improvements in overall simulation time, is easily parallelized, and is scalable in the context of this large-scale computational fluid dynamics application. This approach reduced overall execution time by using cheaper, though less powerful, linear solvers for relatively easy linear systems and then switching to more robust but more costly methods for more difficult linear systems. The results demonstrate that adaptive solvers can be implemented easily in a multiprocessor environment and are scalable. The chapter investigates adaptive solvers in problem domains and considers more adaptive approaches, including a polynomial heuristic where the trends of the indicators can be estimated by fitting a function to known data points. The chapter also combines adaptive heuristics with high-performance component infrastructure for performance monitoring and analysis.",
author = "S. Bhowmick and D. Kaushik and L. McInnes and B. Norris and P. Raghavan",
year = "2006",
month = "12",
day = "1",
doi = "10.1016/B978-044452206-1/50033-1",
language = "English (US)",
isbn = "9780444522061",
pages = "277--284",
booktitle = "Parallel Computational Fluid Dynamics 2005",
publisher = "Elsevier",
address = "United States",

}

TY - CHAP

T1 - Parallel adaptive solvers in compressible petsc-fun3d simulations

AU - Bhowmick, S.

AU - Kaushik, D.

AU - McInnes, L.

AU - Norris, B.

AU - Raghavan, P.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The chapter presents a polyalgorithmic technique for adaptively selecting the linear solver method to match the numeric properties of linear systems as they evolve during the course of nonlinear iterations. The approach combines more robust but more costly methods when needed in particularly challenging phases of solution, with cheaper, though less powerful, methods in other phases. The chapter demonstrates that this adaptive, polyalgorithmic approach leads to improvements in overall simulation time, is easily parallelized, and is scalable in the context of this large-scale computational fluid dynamics application. This approach reduced overall execution time by using cheaper, though less powerful, linear solvers for relatively easy linear systems and then switching to more robust but more costly methods for more difficult linear systems. The results demonstrate that adaptive solvers can be implemented easily in a multiprocessor environment and are scalable. The chapter investigates adaptive solvers in problem domains and considers more adaptive approaches, including a polynomial heuristic where the trends of the indicators can be estimated by fitting a function to known data points. The chapter also combines adaptive heuristics with high-performance component infrastructure for performance monitoring and analysis.

AB - The chapter presents a polyalgorithmic technique for adaptively selecting the linear solver method to match the numeric properties of linear systems as they evolve during the course of nonlinear iterations. The approach combines more robust but more costly methods when needed in particularly challenging phases of solution, with cheaper, though less powerful, methods in other phases. The chapter demonstrates that this adaptive, polyalgorithmic approach leads to improvements in overall simulation time, is easily parallelized, and is scalable in the context of this large-scale computational fluid dynamics application. This approach reduced overall execution time by using cheaper, though less powerful, linear solvers for relatively easy linear systems and then switching to more robust but more costly methods for more difficult linear systems. The results demonstrate that adaptive solvers can be implemented easily in a multiprocessor environment and are scalable. The chapter investigates adaptive solvers in problem domains and considers more adaptive approaches, including a polynomial heuristic where the trends of the indicators can be estimated by fitting a function to known data points. The chapter also combines adaptive heuristics with high-performance component infrastructure for performance monitoring and analysis.

UR - http://www.scopus.com/inward/record.url?scp=84882497048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882497048&partnerID=8YFLogxK

U2 - 10.1016/B978-044452206-1/50033-1

DO - 10.1016/B978-044452206-1/50033-1

M3 - Chapter

AN - SCOPUS:84882497048

SN - 9780444522061

SP - 277

EP - 284

BT - Parallel Computational Fluid Dynamics 2005

PB - Elsevier

ER -