A generic library for large scale solution of PDEs on modern heterogeneous architectures

Publication: Research - peer-reviewConference abstract in proceedings – Annual report year: 2012

Standard

A generic library for large scale solution of PDEs on modern heterogeneous architectures. / Glimberg, Stefan Lemvig; Engsig-Karup, Allan Peter.

7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics, 2012. p. 12.

Publication: Research - peer-reviewConference abstract in proceedings – Annual report year: 2012

Harvard

Glimberg, SL & Engsig-Karup, AP 2012, 'A generic library for large scale solution of PDEs on modern heterogeneous architectures'. in 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics, pp. 12.

APA

Glimberg, S. L., & Engsig-Karup, A. P. (2012). A generic library for large scale solution of PDEs on modern heterogeneous architectures. In 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. (pp. 12). The European Research Consortium for Informatics and Mathematics.

CBE

Glimberg SL, Engsig-Karup AP. 2012. A generic library for large scale solution of PDEs on modern heterogeneous architectures. In 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics. pp. 12.

MLA

Glimberg, Stefan Lemvig and Allan Peter Engsig-Karup "A generic library for large scale solution of PDEs on modern heterogeneous architectures". 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics. 2012. 12.

Vancouver

Glimberg SL, Engsig-Karup AP. A generic library for large scale solution of PDEs on modern heterogeneous architectures. In 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics. 2012. p. 12.

Author

Glimberg, Stefan Lemvig; Engsig-Karup, Allan Peter / A generic library for large scale solution of PDEs on modern heterogeneous architectures.

7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12): Programme and Abstracts. The European Research Consortium for Informatics and Mathematics, 2012. p. 12.

Publication: Research - peer-reviewConference abstract in proceedings – Annual report year: 2012

Bibtex

@inbook{ab2a329428e245969e70a71477f39907,
title = "A generic library for large scale solution of PDEs on modern heterogeneous architectures",
publisher = "The European Research Consortium for Informatics and Mathematics",
author = "Glimberg, {Stefan Lemvig} and Engsig-Karup, {Allan Peter}",
year = "2012",
pages = "12",
booktitle = "7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12)",

}

RIS

TY - ABST

T1 - A generic library for large scale solution of PDEs on modern heterogeneous architectures

A1 - Glimberg,Stefan Lemvig

A1 - Engsig-Karup,Allan Peter

AU - Glimberg,Stefan Lemvig

AU - Engsig-Karup,Allan Peter

PB - The European Research Consortium for Informatics and Mathematics

PY - 2012

Y1 - 2012

N2 - Adapting to new programming models for modern multi- and many-core architectures requires code-rewriting and changing algorithms and data structures, in order to achieve good efficiency and scalability. We present a generic library for solving large scale partial differential equations (PDEs), capable of utilizing heterogeneous CPU/GPU environments. The library can be used for fast proto-typing of PDE solvers, based on finite difference approximations of spatial derivatives in one, two, or three dimensions. In order to efficiently solve large scale problems, we keep memory consumption and memory access low, using a low-storage implementation of flexible-order finite difference operators. We will illustrate the use of library components by assembling such matrix-free operators to be used with one of the supported iterative solvers, such as GMRES, CG, Multigrid or Defect Correction. As a proto-type for large scale PDE solvers, we present the assembling of a tool for simulation of three dimensional fully nonlinear water waves. Measurements show scalable performance results - in the same order as a dedicated non-library version of the wave tool. Introducing a domain decomposition preconditioner based on a multigrid method, further extends support for multiple GPUs and allows for improvements in performance as well as increased problem sizes.

AB - Adapting to new programming models for modern multi- and many-core architectures requires code-rewriting and changing algorithms and data structures, in order to achieve good efficiency and scalability. We present a generic library for solving large scale partial differential equations (PDEs), capable of utilizing heterogeneous CPU/GPU environments. The library can be used for fast proto-typing of PDE solvers, based on finite difference approximations of spatial derivatives in one, two, or three dimensions. In order to efficiently solve large scale problems, we keep memory consumption and memory access low, using a low-storage implementation of flexible-order finite difference operators. We will illustrate the use of library components by assembling such matrix-free operators to be used with one of the supported iterative solvers, such as GMRES, CG, Multigrid or Defect Correction. As a proto-type for large scale PDE solvers, we present the assembling of a tool for simulation of three dimensional fully nonlinear water waves. Measurements show scalable performance results - in the same order as a dedicated non-library version of the wave tool. Introducing a domain decomposition preconditioner based on a multigrid method, further extends support for multiple GPUs and allows for improvements in performance as well as increased problem sizes.

BT - 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12)

T2 - 7th International Workshop on Parallel Matrix Algorithms and Applications (PMAA’12)

SP - 12

ER -