Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

Svetlana Nogina; Kristof Unterweger; Tobias Weinzierl

doi:10.1007/978-3-642-31464-3_68

Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

Svetlana Nogina, Kristof Unterweger, Tobias Weinzierl

Research output: Chapter in Book/Report/Conference proceeding › Chapter

4 Scopus citations

Abstract

Many multithreaded, grid-based, dynamically adaptive solvers for partial differential equations permanently have to traverse subgrids (patches) of different and changing sizes. The parallel efficiency of this traversal depends on the interplay of the patch size, the architecture used, the operations triggered throughout the traversal, and the grain size, i.e. the size of the subtasks the patch is broken into. We propose an oracle mechanism delivering grain sizes on-the-fly. It takes historical runtime measurements for different patch and grain sizes as well as the traverse's operations into account, and it yields reasonable speedups. Neither magic configuration settings nor an expensive pre-tuning phase are necessary. It is an autotuning approach. © 2012 Springer-Verlag.

Original language	English (US)
Title of host publication	Parallel Processing and Applied Mathematics
Publisher	Springer Nature
Pages	671-680
Number of pages	10
ISBN (Print)	9783642314636
DOIs	https://doi.org/10.1007/978-3-642-31464-3_68
State	Published - 2012
Externally published	Yes

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): UK-c0020
Acknowledgements: This publication is partially based on work supportedby Award No. UK-c0020, made by the King Abdullah University of Science andTechnology (KAUST). Computing resources for the present work have also beenprovided by the Gauss Centre for Supercomputing under grant pr63no.
This publication acknowledges KAUST support, but has no KAUST affiliated authors.

Access to Document

10.1007/978-3-642-31464-3_68

Cite this

@inbook{225c0e9ce09b4af49f50e21e22a7e905,

title = "Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures",

abstract = "Many multithreaded, grid-based, dynamically adaptive solvers for partial differential equations permanently have to traverse subgrids (patches) of different and changing sizes. The parallel efficiency of this traversal depends on the interplay of the patch size, the architecture used, the operations triggered throughout the traversal, and the grain size, i.e. the size of the subtasks the patch is broken into. We propose an oracle mechanism delivering grain sizes on-the-fly. It takes historical runtime measurements for different patch and grain sizes as well as the traverse's operations into account, and it yields reasonable speedups. Neither magic configuration settings nor an expensive pre-tuning phase are necessary. It is an autotuning approach. {\textcopyright} 2012 Springer-Verlag.",

author = "Svetlana Nogina and Kristof Unterweger and Tobias Weinzierl",

note = "KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): UK-c0020 Acknowledgements: This publication is partially based on work supportedby Award No. UK-c0020, made by the King Abdullah University of Science andTechnology (KAUST). Computing resources for the present work have also beenprovided by the Gauss Centre for Supercomputing under grant pr63no. This publication acknowledges KAUST support, but has no KAUST affiliated authors.",

year = "2012",

doi = "10.1007/978-3-642-31464-3_68",

language = "English (US)",

isbn = "9783642314636",

pages = "671--680",

booktitle = "Parallel Processing and Applied Mathematics",

publisher = "Springer Nature",

}

TY - CHAP

T1 - Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

AU - Nogina, Svetlana

AU - Unterweger, Kristof

AU - Weinzierl, Tobias

N1 - KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): UK-c0020 Acknowledgements: This publication is partially based on work supportedby Award No. UK-c0020, made by the King Abdullah University of Science andTechnology (KAUST). Computing resources for the present work have also beenprovided by the Gauss Centre for Supercomputing under grant pr63no. This publication acknowledges KAUST support, but has no KAUST affiliated authors.

PY - 2012

Y1 - 2012

N2 - Many multithreaded, grid-based, dynamically adaptive solvers for partial differential equations permanently have to traverse subgrids (patches) of different and changing sizes. The parallel efficiency of this traversal depends on the interplay of the patch size, the architecture used, the operations triggered throughout the traversal, and the grain size, i.e. the size of the subtasks the patch is broken into. We propose an oracle mechanism delivering grain sizes on-the-fly. It takes historical runtime measurements for different patch and grain sizes as well as the traverse's operations into account, and it yields reasonable speedups. Neither magic configuration settings nor an expensive pre-tuning phase are necessary. It is an autotuning approach. © 2012 Springer-Verlag.

AB - Many multithreaded, grid-based, dynamically adaptive solvers for partial differential equations permanently have to traverse subgrids (patches) of different and changing sizes. The parallel efficiency of this traversal depends on the interplay of the patch size, the architecture used, the operations triggered throughout the traversal, and the grain size, i.e. the size of the subtasks the patch is broken into. We propose an oracle mechanism delivering grain sizes on-the-fly. It takes historical runtime measurements for different patch and grain sizes as well as the traverse's operations into account, and it yields reasonable speedups. Neither magic configuration settings nor an expensive pre-tuning phase are necessary. It is an autotuning approach. © 2012 Springer-Verlag.

UR - http://hdl.handle.net/10754/597640

UR - http://link.springer.com/10.1007/978-3-642-31464-3_68

UR - http://www.scopus.com/inward/record.url?scp=84865210278&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-31464-3_68

DO - 10.1007/978-3-642-31464-3_68

M3 - Chapter

SN - 9783642314636

SP - 671

EP - 680

BT - Parallel Processing and Applied Mathematics

PB - Springer Nature

ER -

Autotuning of Adaptive Mesh Refinement PDE Solvers on Shared Memory Architectures

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this