Computing least squares condition numbers on hybrid multicore/GPU systems

M. Baboulin; J. Dongarra; R. Lacroix

doi:10.1007/978-3-319-12307-3_6

Computing least squares condition numbers on hybrid multicore/GPU systems

M. Baboulin
, J. Dongarra
, R. Lacroix

Computer Science and Mathematics Div

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

1 Scopus citations

Abstract

This chapter presents an efficient computation for least squares conditioning or estimates of it. We propose performance results using new routines on top of the multicore-GPU library MAGMA. This set of routines is based on an efficient computation of the variance–covariance matrix for which, to our knowledge, there is no implementation in current public domain libraries LAPACK and ScaLAPACK.

Original language	English
Title of host publication	Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science
Editors	Roman N. Makarov, Roderick V. N. Melnik, Ilias S. Kotsireas, Hasan Shodiev, Monica G. Cojocaru, Monica G. Cojocaru, Roman N. Makarov, Roderick V. N. Melnik, Ilias S. Kotsireas, Hasan Shodiev
Publisher	Springer New York LLC
Pages	35-41
Number of pages	7
ISBN (Print)	9783319123066, 9783319123066
DOIs	https://doi.org/10.1007/978-3-319-12307-3_6
State	Published - 2015
Event	International Conference on Applied Mathematics, Modelling and Computational Science, AMMCS 2013 - Waterloo, Canada Duration: Aug 26 2013 → Aug 30 2013

Publication series

Name	Springer Proceedings in Mathematics and Statistics
Volume	117
ISSN (Print)	2194-1009
ISSN (Electronic)	2194-1017

Conference

Conference	International Conference on Applied Mathematics, Modelling and Computational Science, AMMCS 2013
Country/Territory	Canada
City	Waterloo
Period	08/26/13 → 08/30/13

Access to Document

10.1007/978-3-319-12307-3_6

Cite this

Baboulin, M., Dongarra, J., & Lacroix, R. (2015). Computing least squares condition numbers on hybrid multicore/GPU systems. In R. N. Makarov, R. V. N. Melnik, I. S. Kotsireas, H. Shodiev, M. G. Cojocaru, M. G. Cojocaru, R. N. Makarov, R. V. N. Melnik, I. S. Kotsireas, & H. Shodiev (Eds.), Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science (pp. 35-41). (Springer Proceedings in Mathematics and Statistics; Vol. 117). Springer New York LLC. https://doi.org/10.1007/978-3-319-12307-3_6

Baboulin, M. ; Dongarra, J. ; Lacroix, R. / Computing least squares condition numbers on hybrid multicore/GPU systems. Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science. editor / Roman N. Makarov ; Roderick V. N. Melnik ; Ilias S. Kotsireas ; Hasan Shodiev ; Monica G. Cojocaru ; Monica G. Cojocaru ; Roman N. Makarov ; Roderick V. N. Melnik ; Ilias S. Kotsireas ; Hasan Shodiev. Springer New York LLC, 2015. pp. 35-41 (Springer Proceedings in Mathematics and Statistics).

@inproceedings{ebf1abab5198484f9f27b10219550dc7,

title = "Computing least squares condition numbers on hybrid multicore/GPU systems",

abstract = "This chapter presents an efficient computation for least squares conditioning or estimates of it. We propose performance results using new routines on top of the multicore-GPU library MAGMA. This set of routines is based on an efficient computation of the variance{\textendash}covariance matrix for which, to our knowledge, there is no implementation in current public domain libraries LAPACK and ScaLAPACK.",

author = "M. Baboulin and J. Dongarra and R. Lacroix",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; International Conference on Applied Mathematics, Modelling and Computational Science, AMMCS 2013 ; Conference date: 26-08-2013 Through 30-08-2013",

year = "2015",

doi = "10.1007/978-3-319-12307-3\_6",

language = "English",

isbn = "9783319123066",

series = "Springer Proceedings in Mathematics and Statistics",

publisher = "Springer New York LLC",

pages = "35--41",

editor = "Makarov, \{Roman N.\} and Melnik, \{Roderick V. N.\} and Kotsireas, \{Ilias S.\} and Hasan Shodiev and Cojocaru, \{Monica G.\} and Cojocaru, \{Monica G.\} and Makarov, \{Roman N.\} and Melnik, \{Roderick V. N.\} and Kotsireas, \{Ilias S.\} and Hasan Shodiev",

booktitle = "Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science",

}

Baboulin, M, Dongarra, J & Lacroix, R 2015, Computing least squares condition numbers on hybrid multicore/GPU systems. in RN Makarov, RVN Melnik, IS Kotsireas, H Shodiev, MG Cojocaru, MG Cojocaru, RN Makarov, RVN Melnik, IS Kotsireas & H Shodiev (eds), Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science. Springer Proceedings in Mathematics and Statistics, vol. 117, Springer New York LLC, pp. 35-41, International Conference on Applied Mathematics, Modelling and Computational Science, AMMCS 2013, Waterloo, Canada, 08/26/13. https://doi.org/10.1007/978-3-319-12307-3_6

Computing least squares condition numbers on hybrid multicore/GPU systems. / Baboulin, M.; Dongarra, J.; Lacroix, R.
Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science. ed. / Roman N. Makarov; Roderick V. N. Melnik; Ilias S. Kotsireas; Hasan Shodiev; Monica G. Cojocaru; Monica G. Cojocaru; Roman N. Makarov; Roderick V. N. Melnik; Ilias S. Kotsireas; Hasan Shodiev. Springer New York LLC, 2015. p. 35-41 (Springer Proceedings in Mathematics and Statistics; Vol. 117).

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

TY - GEN

T1 - Computing least squares condition numbers on hybrid multicore/GPU systems

AU - Baboulin, M.

AU - Dongarra, J.

AU - Lacroix, R.

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - This chapter presents an efficient computation for least squares conditioning or estimates of it. We propose performance results using new routines on top of the multicore-GPU library MAGMA. This set of routines is based on an efficient computation of the variance–covariance matrix for which, to our knowledge, there is no implementation in current public domain libraries LAPACK and ScaLAPACK.

AB - This chapter presents an efficient computation for least squares conditioning or estimates of it. We propose performance results using new routines on top of the multicore-GPU library MAGMA. This set of routines is based on an efficient computation of the variance–covariance matrix for which, to our knowledge, there is no implementation in current public domain libraries LAPACK and ScaLAPACK.

UR - https://www.scopus.com/pages/publications/84951174133

U2 - 10.1007/978-3-319-12307-3_6

DO - 10.1007/978-3-319-12307-3_6

M3 - Conference contribution

AN - SCOPUS:84951174133

SN - 9783319123066

T3 - Springer Proceedings in Mathematics and Statistics

SP - 35

EP - 41

BT - Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science

A2 - Makarov, Roman N.

A2 - Melnik, Roderick V. N.

A2 - Kotsireas, Ilias S.

A2 - Shodiev, Hasan

A2 - Cojocaru, Monica G.

A2 - Makarov, Roman N.

A2 - Melnik, Roderick V. N.

A2 - Kotsireas, Ilias S.

A2 - Shodiev, Hasan

PB - Springer New York LLC

T2 - International Conference on Applied Mathematics, Modelling and Computational Science, AMMCS 2013

Y2 - 26 August 2013 through 30 August 2013

ER -

Baboulin M, Dongarra J, Lacroix R. Computing least squares condition numbers on hybrid multicore/GPU systems. In Makarov RN, Melnik RVN, Kotsireas IS, Shodiev H, Cojocaru MG, Cojocaru MG, Makarov RN, Melnik RVN, Kotsireas IS, Shodiev H, editors, Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science. Springer New York LLC. 2015. p. 35-41. (Springer Proceedings in Mathematics and Statistics). doi: 10.1007/978-3-319-12307-3_6

Computing least squares condition numbers on hybrid multicore/GPU systems

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this