Implementing the conjugate gradient algorithm on multi-core systems

W.A. Wiggers; V. Bakker; A.B.J. Kokkeler; G.J.M. Smit

doi:10.1109/ISSOC.2007.4427436

Implementing the conjugate gradient algorithm on multi-core systems

W.A. Wiggers, V. Bakker, A.B.J. Kokkeler, G.J.M. Smit

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

19 Citations (Scopus)

185 Downloads (Pure)

Abstract

In linear solvers, like the conjugate gradient algorithm, sparse-matrix vector multiplication is an important kernel. Due to the sparseness of the matrices, the solver runs relatively slow. For digital optical tomography (DOT), a large set of linear equations have to be solved which currently takes in the order of hours on desktop computers. Our goal was to speed up the conjugate gradient solver. In this paper we present the results of applying multiple optimization techniques and exploiting multi-core solutions offered by two recently introduced architectures: Intel’s Woodcrest general purpose processor and NVIDIA’s G80 graphical processing unit. Using these techniques for these architectures, a speedup of a factor three has been achieved.

Original language	English
Title of host publication	Proceedings of the International Symposium on System-on-Chip (SoC 2007)
Editors	J. Nurmi, J. Takala, O. Vainio
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	11-14
Number of pages	4
ISBN (Print)	978-1-4244-1367-6, 978-1-4244-1368-3 (CD)
DOIs	https://doi.org/10.1109/ISSOC.2007.4427436
Publication status	Published - 20 Nov 2007
Event	International Symposium on System-on-Chip, SoC 2007 - Tampere, Finland Duration: 19 Nov 2007 → 21 Nov 2007

Conference

Conference	International Symposium on System-on-Chip, SoC 2007
Abbreviated title	SoC
Country/Territory	Finland
City	Tampere
Period	19/11/07 → 21/11/07

Keywords

CAES-EEA: Efficient Embedded Architectures

Access to Document

10.1109/ISSOC.2007.4427436

ArticleAccepted author version, 106 KB

Cite this

@inproceedings{d4171de957ac49dba2e765e8b16edf3f,

title = "Implementing the conjugate gradient algorithm on multi-core systems",

abstract = "In linear solvers, like the conjugate gradient algorithm, sparse-matrix vector multiplication is an important kernel. Due to the sparseness of the matrices, the solver runs relatively slow. For digital optical tomography (DOT), a large set of linear equations have to be solved which currently takes in the order of hours on desktop computers. Our goal was to speed up the conjugate gradient solver. In this paper we present the results of applying multiple optimization techniques and exploiting multi-core solutions offered by two recently introduced architectures: Intel{\textquoteright}s Woodcrest general purpose processor and NVIDIA{\textquoteright}s G80 graphical processing unit. Using these techniques for these architectures, a speedup of a factor three has been achieved.",

keywords = "CAES-EEA: Efficient Embedded Architectures",

author = "W.A. Wiggers and V. Bakker and A.B.J. Kokkeler and G.J.M. Smit",

year = "2007",

month = nov,

day = "20",

doi = "10.1109/ISSOC.2007.4427436",

language = "English",

isbn = "978-1-4244-1367-6",

pages = "11--14",

editor = "J. Nurmi and J. Takala and O. Vainio",

booktitle = "Proceedings of the International Symposium on System-on-Chip (SoC 2007)",

publisher = "IEEE",

address = "United States",

note = "International Symposium on System-on-Chip, SoC 2007, SoC ; Conference date: 19-11-2007 Through 21-11-2007",

}

Wiggers, WA, Bakker, V, Kokkeler, ABJ & Smit, GJM 2007, Implementing the conjugate gradient algorithm on multi-core systems. in J Nurmi, J Takala & O Vainio (eds), Proceedings of the International Symposium on System-on-Chip (SoC 2007). IEEE, Piscataway, NJ, pp. 11-14, International Symposium on System-on-Chip, SoC 2007, Tampere, Finland, 19/11/07. https://doi.org/10.1109/ISSOC.2007.4427436

Implementing the conjugate gradient algorithm on multi-core systems. / Wiggers, W.A.; Bakker, V.; Kokkeler, A.B.J. et al.
Proceedings of the International Symposium on System-on-Chip (SoC 2007). ed. / J. Nurmi; J. Takala; O. Vainio. Piscataway, NJ: IEEE, 2007. p. 11-14.

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

TY - GEN

T1 - Implementing the conjugate gradient algorithm on multi-core systems

AU - Wiggers, W.A.

AU - Bakker, V.

AU - Kokkeler, A.B.J.

AU - Smit, G.J.M.

PY - 2007/11/20

Y1 - 2007/11/20

N2 - In linear solvers, like the conjugate gradient algorithm, sparse-matrix vector multiplication is an important kernel. Due to the sparseness of the matrices, the solver runs relatively slow. For digital optical tomography (DOT), a large set of linear equations have to be solved which currently takes in the order of hours on desktop computers. Our goal was to speed up the conjugate gradient solver. In this paper we present the results of applying multiple optimization techniques and exploiting multi-core solutions offered by two recently introduced architectures: Intel’s Woodcrest general purpose processor and NVIDIA’s G80 graphical processing unit. Using these techniques for these architectures, a speedup of a factor three has been achieved.

AB - In linear solvers, like the conjugate gradient algorithm, sparse-matrix vector multiplication is an important kernel. Due to the sparseness of the matrices, the solver runs relatively slow. For digital optical tomography (DOT), a large set of linear equations have to be solved which currently takes in the order of hours on desktop computers. Our goal was to speed up the conjugate gradient solver. In this paper we present the results of applying multiple optimization techniques and exploiting multi-core solutions offered by two recently introduced architectures: Intel’s Woodcrest general purpose processor and NVIDIA’s G80 graphical processing unit. Using these techniques for these architectures, a speedup of a factor three has been achieved.

KW - CAES-EEA: Efficient Embedded Architectures

U2 - 10.1109/ISSOC.2007.4427436

DO - 10.1109/ISSOC.2007.4427436

M3 - Conference contribution

SN - 978-1-4244-1367-6

SN - 978-1-4244-1368-3 (CD)

SP - 11

EP - 14

BT - Proceedings of the International Symposium on System-on-Chip (SoC 2007)

A2 - Nurmi, J.

A2 - Takala, J.

A2 - Vainio, O.

PB - IEEE

CY - Piscataway, NJ

T2 - International Symposium on System-on-Chip, SoC 2007

Y2 - 19 November 2007 through 21 November 2007

ER -

Implementing the conjugate gradient algorithm on multi-core systems

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this