Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs

Hartwig Anzt; Moritz Kreutzer; Eduardo Ponce; Gregory Peterson; Gerhard Wellein; Jack Dongarra

doi:10.1177/1094342016646844

Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs

Hartwig Anzt, Moritz Kreutzer, Eduardo Ponce, Gregory Peterson, Gerhard Wellein, Jack Dongarra

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

Abstract

In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.

Original language	English
Pages (from-to)	1-11
Journal	International Journal of High Performance Computing Applications
Early online date	5 May 2016
DOIs	https://doi.org/10.1177/1094342016646844
Publication status	Published - 2016

Access to Document

10.1177/1094342016646844

Cite this

@article{46d836fb497b420e99e78775da8f19ac,

title = "Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs",

abstract = "In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.",

author = "Hartwig Anzt and Moritz Kreutzer and Eduardo Ponce and Gregory Peterson and Gerhard Wellein and Jack Dongarra",

year = "2016",

doi = "10.1177/1094342016646844",

language = "English",

pages = "1--11",

journal = "International Journal of High Performance Computing Applications",

issn = "1094-3420",

publisher = "Sage Publications Ltd",

}

TY - JOUR

T1 - Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs

AU - Anzt, Hartwig

AU - Kreutzer, Moritz

AU - Ponce, Eduardo

AU - Peterson, Gregory

AU - Wellein, Gerhard

AU - Dongarra, Jack

PY - 2016

Y1 - 2016

N2 - In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.

AB - In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.

U2 - 10.1177/1094342016646844

DO - 10.1177/1094342016646844

M3 - Article

SN - 1094-3420

SP - 1

EP - 11

JO - International Journal of High Performance Computing Applications

JF - International Journal of High Performance Computing Applications

ER -

Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs

Abstract

Access to Document

Fingerprint

Cite this