Batched generation of incomplete sparse approximate inverses on GPUs

Hartwig Anzt; Edmond Chow; Thomas Huckle; Jack Dongarra

doi:10.1109/ScalA.2016.011

Batched generation of incomplete sparse approximate inverses on GPUs

Hartwig Anzt, Edmond Chow, Thomas Huckle, Jack Dongarra

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

16 Scopus citations

Abstract

Incomplete Sparse Approximate Inverses (ISAI) have recently been shown to be an attractive alternative to exact sparse triangular solves in the context of incomplete factorization preconditioning. In this paper we propose a batched GPU-kernel for the efficient generation of ISAI matrices. Utilizing only thread-local memory allows for computing the ISAI matrix with very small memory footprint. We demonstrate that this strategy is faster than the existing strategy for generating ISAI matrices, and use a large number of test matrices to assess the algorithm's efficiency in an iterative solver setting.

Original language	English
Title of host publication	Proceedings of ScalA 2016
Subtitle of host publication	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	49-56
Number of pages	8
ISBN (Electronic)	9781509052226
DOIs	https://doi.org/10.1109/ScalA.2016.011
State	Published - Jan 30 2017
Externally published	Yes
Event	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016 - Salt Lake City, United States Duration: Nov 13 2016 → Nov 18 2016

Publication series

Name	Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016
Country/Territory	United States
City	Salt Lake City
Period	11/13/16 → 11/18/16

Funding

This material is based upon work supported by the U.S. Department of Energy Office of Science, Office of Advanced Scientific Computing Research, Applied Mathematics program under Award Numbers DE-SC0016564 and DE-SC0016513, and NVIDIA.

Keywords

Batched routines
GPU
Incomplete Sparse Approximate Inverses
Preconditioning

Access to Document

10.1109/ScalA.2016.011

Cite this

Anzt, H., Chow, E., Huckle, T., & Dongarra, J. (2017). Batched generation of incomplete sparse approximate inverses on GPUs. In Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 49-56). Article 7836596 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ScalA.2016.011

Anzt, Hartwig ; Chow, Edmond ; Huckle, Thomas et al. / Batched generation of incomplete sparse approximate inverses on GPUs. Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 49-56 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{a0332c903d8b49bd8af377b06517a395,

title = "Batched generation of incomplete sparse approximate inverses on GPUs",

abstract = "Incomplete Sparse Approximate Inverses (ISAI) have recently been shown to be an attractive alternative to exact sparse triangular solves in the context of incomplete factorization preconditioning. In this paper we propose a batched GPU-kernel for the efficient generation of ISAI matrices. Utilizing only thread-local memory allows for computing the ISAI matrix with very small memory footprint. We demonstrate that this strategy is faster than the existing strategy for generating ISAI matrices, and use a large number of test matrices to assess the algorithm's efficiency in an iterative solver setting.",

keywords = "Batched routines, GPU, Incomplete Sparse Approximate Inverses, Preconditioning",

author = "Hartwig Anzt and Edmond Chow and Thomas Huckle and Jack Dongarra",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016 ; Conference date: 13-11-2016 Through 18-11-2016",

year = "2017",

month = jan,

day = "30",

doi = "10.1109/ScalA.2016.011",

language = "English",

series = "Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "49--56",

booktitle = "Proceedings of ScalA 2016",

}

Anzt, H, Chow, E, Huckle, T & Dongarra, J 2017, Batched generation of incomplete sparse approximate inverses on GPUs. in Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis., 7836596, Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 49-56, 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016, Salt Lake City, United States, 11/13/16. https://doi.org/10.1109/ScalA.2016.011

Batched generation of incomplete sparse approximate inverses on GPUs. / Anzt, Hartwig; Chow, Edmond; Huckle, Thomas et al.
Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. p. 49-56 7836596 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Batched generation of incomplete sparse approximate inverses on GPUs

AU - Anzt, Hartwig

AU - Chow, Edmond

AU - Huckle, Thomas

AU - Dongarra, Jack

PY - 2017/1/30

Y1 - 2017/1/30

N2 - Incomplete Sparse Approximate Inverses (ISAI) have recently been shown to be an attractive alternative to exact sparse triangular solves in the context of incomplete factorization preconditioning. In this paper we propose a batched GPU-kernel for the efficient generation of ISAI matrices. Utilizing only thread-local memory allows for computing the ISAI matrix with very small memory footprint. We demonstrate that this strategy is faster than the existing strategy for generating ISAI matrices, and use a large number of test matrices to assess the algorithm's efficiency in an iterative solver setting.

AB - Incomplete Sparse Approximate Inverses (ISAI) have recently been shown to be an attractive alternative to exact sparse triangular solves in the context of incomplete factorization preconditioning. In this paper we propose a batched GPU-kernel for the efficient generation of ISAI matrices. Utilizing only thread-local memory allows for computing the ISAI matrix with very small memory footprint. We demonstrate that this strategy is faster than the existing strategy for generating ISAI matrices, and use a large number of test matrices to assess the algorithm's efficiency in an iterative solver setting.

KW - Batched routines

KW - GPU

KW - Incomplete Sparse Approximate Inverses

KW - Preconditioning

UR - http://www.scopus.com/inward/record.url?scp=85015226986&partnerID=8YFLogxK

U2 - 10.1109/ScalA.2016.011

DO - 10.1109/ScalA.2016.011

M3 - Conference contribution

AN - SCOPUS:85015226986

T3 - Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 49

EP - 56

BT - Proceedings of ScalA 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016

Y2 - 13 November 2016 through 18 November 2016

ER -

Anzt H, Chow E, Huckle T, Dongarra J. Batched generation of incomplete sparse approximate inverses on GPUs. In Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2017. p. 49-56. 7836596. (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/ScalA.2016.011

Batched generation of incomplete sparse approximate inverses on GPUs

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this