Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices

Karin Meyer; Bruce Tier

Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices

Karin Meyer, Bruce Tier

Animal Genetics and Breeding Unit

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

Abstract

The era of genomic evaluation has brought the need to perform computations involving large, dense matrices. Particular tasks are the computation and inversion of the genomic relationship matrix. This paper investigates the suitability of Graphics Processing Units together with highly optimised software libraries for these computations, using blocked algorithms. It is shown that calculations are readily sped up by parallel processing, using freely available library routines, and that reductions in time by factors of 4 to 5 are achievable even for 'consumer' grade graphics cards.

Original language	English
Title of host publication	Proceedings of the Association for the Advancement of Animal Breeding and Genetics
Editors	Nicolas Lopez Villalobos
Place of Publication	Armidale, Australia
Publisher	Association for the Advancement of Animal Breeding and Genetics (AAABG)
Pages	270-273
Volume	20
ISBN (Print)	9780473260569
Publication status	Published - 2013
Event	AAABG 2013: 20th Conference of the Association for the Advancement of Animal Breeding and Genetics: Translating Science into Action - Napier, New Zealand Duration: 20 Oct 2013 → 23 Oct 2013

Conference

Conference	AAABG 2013: 20th Conference of the Association for the Advancement of Animal Breeding and Genetics: Translating Science into Action
City	Napier, New Zealand
Period	20/10/13 → 23/10/13

Keywords

Genetics

Access to Document

https://hdl.handle.net/1959.11/14323

Handle.net

Persistent link

Cite this

Meyer, K., & Tier, B. (2013). Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices. In N. Lopez Villalobos (Ed.), Proceedings of the Association for the Advancement of Animal Breeding and Genetics (Vol. 20, pp. 270-273). Association for the Advancement of Animal Breeding and Genetics (AAABG). https://hdl.handle.net/1959.11/14323

Meyer, Karin ; Tier, Bruce. / Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices. Proceedings of the Association for the Advancement of Animal Breeding and Genetics. editor / Nicolas Lopez Villalobos. Vol. 20 Armidale, Australia : Association for the Advancement of Animal Breeding and Genetics (AAABG), 2013. pp. 270-273

@inproceedings{999bbe9e7fc743c6a57f6e7e2464d234,

title = "Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices",

abstract = "The era of genomic evaluation has brought the need to perform computations involving large, dense matrices. Particular tasks are the computation and inversion of the genomic relationship matrix. This paper investigates the suitability of Graphics Processing Units together with highly optimised software libraries for these computations, using blocked algorithms. It is shown that calculations are readily sped up by parallel processing, using freely available library routines, and that reductions in time by factors of 4 to 5 are achievable even for 'consumer' grade graphics cards.",

keywords = "Genetics",

author = "Karin Meyer and Bruce Tier",

year = "2013",

language = "English",

isbn = "9780473260569",

volume = "20",

pages = "270--273",

editor = "\{Lopez Villalobos\}, Nicolas",

booktitle = "Proceedings of the Association for the Advancement of Animal Breeding and Genetics",

publisher = "Association for the Advancement of Animal Breeding and Genetics (AAABG)",

note = "AAABG 2013: 20th Conference of the Association for the Advancement of Animal Breeding and Genetics: Translating Science into Action ; Conference date: 20-10-2013 Through 23-10-2013",

}

Meyer, K & Tier, B 2013, Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices. in N Lopez Villalobos (ed.), Proceedings of the Association for the Advancement of Animal Breeding and Genetics. vol. 20, Association for the Advancement of Animal Breeding and Genetics (AAABG), Armidale, Australia, pp. 270-273, AAABG 2013: 20th Conference of the Association for the Advancement of Animal Breeding and Genetics: Translating Science into Action, Napier, New Zealand, 20/10/13. <https://hdl.handle.net/1959.11/14323>

Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices. / Meyer, Karin; Tier, Bruce.
Proceedings of the Association for the Advancement of Animal Breeding and Genetics. ed. / Nicolas Lopez Villalobos. Vol. 20 Armidale, Australia: Association for the Advancement of Animal Breeding and Genetics (AAABG), 2013. p. 270-273.

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

TY - GEN

T1 - Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices

AU - Meyer, Karin

AU - Tier, Bruce

PY - 2013

Y1 - 2013

N2 - The era of genomic evaluation has brought the need to perform computations involving large, dense matrices. Particular tasks are the computation and inversion of the genomic relationship matrix. This paper investigates the suitability of Graphics Processing Units together with highly optimised software libraries for these computations, using blocked algorithms. It is shown that calculations are readily sped up by parallel processing, using freely available library routines, and that reductions in time by factors of 4 to 5 are achievable even for 'consumer' grade graphics cards.

AB - The era of genomic evaluation has brought the need to perform computations involving large, dense matrices. Particular tasks are the computation and inversion of the genomic relationship matrix. This paper investigates the suitability of Graphics Processing Units together with highly optimised software libraries for these computations, using blocked algorithms. It is shown that calculations are readily sped up by parallel processing, using freely available library routines, and that reductions in time by factors of 4 to 5 are achievable even for 'consumer' grade graphics cards.

KW - Genetics

UR - http://www.aaabg.org/aaabghome/proceedings20.php

M3 - Conference contribution

SN - 9780473260569

VL - 20

SP - 270

EP - 273

BT - Proceedings of the Association for the Advancement of Animal Breeding and Genetics

A2 - Lopez Villalobos, Nicolas

PB - Association for the Advancement of Animal Breeding and Genetics (AAABG)

CY - Armidale, Australia

T2 - AAABG 2013: 20th Conference of the Association for the Advancement of Animal Breeding and Genetics: Translating Science into Action

Y2 - 20 October 2013 through 23 October 2013

ER -

Meyer K, Tier B. Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices. In Lopez Villalobos N, editor, Proceedings of the Association for the Advancement of Animal Breeding and Genetics. Vol. 20. Armidale, Australia: Association for the Advancement of Animal Breeding and Genetics (AAABG). 2013. p. 270-273

Utility of Graphics Processing Units for Dense Matrix Calculations in Computing and Inverting Genomic Relationship Matrices

Abstract

Conference

Keywords

Access to Document

Handle.net

Fingerprint

Cite this