Accelerating High-Throughput Computing through OpenCL

Andrei- Dafinoiu; Joshua Higgins; Violeta Holmes

Accelerating High-Throughput Computing through OpenCL

Andrei- Dafinoiu, Joshua Higgins, Violeta Holmes

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

Abstract

As the computational trend diverges from standard CPU computing, to encompass GPUs and other accelerators, the need to integrate these unused resources within existing systems becomes apparent. This paper presents the implementation of an HTCondor pool with GPU execution capabilities through OpenCL. Implementation is discussed from both the system setup and the software design standpoint. The GPU landscape is investigated and the efficiency of the system is evaluated via FastFourier Transform computations. Experimental results show that HTCondor GPU performance matches a dedicated GPU cluster.

Original language	English
Title of host publication	Proceedings of the EMerging Technology (EMiT) Conference 2016
Editors	B.D. Rogers, D. Topping, F. Mantovani, M.K. Bane
Place of Publication	Manchester
Publisher	The Emerging Technology (EMiT) Conference
Pages	46-49
Number of pages	4
ISBN (Electronic)	9780993342639
Publication status	Published - 3 Jun 2016
Event	Emerging Technology Conference: Emerging Technology: Revolutionising Computing! - Barcelona Supercomputing Center, Barcelona, Spain Duration: 2 Jun 2016 → 3 Jun 2016 https://emit.tech/emit-2016/emit2016-intro/ (Link to Conference Website )

Conference

Conference	Emerging Technology Conference
Abbreviated title	EMiT 2016
Country/Territory	Spain
City	Barcelona
Period	2/06/16 → 3/06/16
Internet address	https://emit.tech/emit-2016/emit2016-intro/ (Link to Conference Website )

Access to Document

http://eprints.hud.ac.uk/id/eprint/28627/Licence: Unspecified
https://emit.tech/EMiT2016/EMiT2016_proceedings.pdfLicence: Unspecified

Cite this

@inproceedings{b4a79aabef33437abf960fb79132da4f,

title = "Accelerating High-Throughput Computing through OpenCL",

abstract = "As the computational trend diverges from standard CPU computing, to encompass GPUs and other accelerators, the need to integrate these unused resources within existing systems becomes apparent. This paper presents the implementation of an HTCondor pool with GPU execution capabilities through OpenCL. Implementation is discussed from both the system setup and the software design standpoint. The GPU landscape is investigated and the efficiency of the system is evaluated via FastFourier Transform computations. Experimental results show that HTCondor GPU performance matches a dedicated GPU cluster.",

keywords = "HPC, HTCondor, OpenCL, GPU",

author = "Andrei- Dafinoiu and Joshua Higgins and Violeta Holmes",

year = "2016",

month = jun,

day = "3",

language = "English",

pages = "46--49",

editor = "B.D. Rogers and D. Topping and F. Mantovani and M.K. Bane",

booktitle = "Proceedings of the EMerging Technology (EMiT) Conference 2016",

publisher = "The Emerging Technology (EMiT) Conference",

address = "United Kingdom",

note = "Emerging Technology Conference : Emerging Technology: Revolutionising Computing!, EMiT 2016 ; Conference date: 02-06-2016 Through 03-06-2016",

url = "https://emit.tech/emit-2016/emit2016-intro/",

}

Dafinoiu, A, Higgins, J & Holmes, V 2016, Accelerating High-Throughput Computing through OpenCL. in BD Rogers, D Topping, F Mantovani & MK Bane (eds), Proceedings of the EMerging Technology (EMiT) Conference 2016. The Emerging Technology (EMiT) Conference, Manchester, pp. 46-49, Emerging Technology Conference, Barcelona, Spain, 2/06/16. <http://eprints.hud.ac.uk/id/eprint/28627/>

Accelerating High-Throughput Computing through OpenCL. / Dafinoiu, Andrei-; Higgins, Joshua; Holmes, Violeta.

Proceedings of the EMerging Technology (EMiT) Conference 2016. ed. / B.D. Rogers; D. Topping; F. Mantovani; M.K. Bane. Manchester : The Emerging Technology (EMiT) Conference, 2016. p. 46-49.

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

TY - GEN

T1 - Accelerating High-Throughput Computing through OpenCL

AU - Dafinoiu, Andrei-

AU - Higgins, Joshua

AU - Holmes, Violeta

PY - 2016/6/3

Y1 - 2016/6/3

N2 - As the computational trend diverges from standard CPU computing, to encompass GPUs and other accelerators, the need to integrate these unused resources within existing systems becomes apparent. This paper presents the implementation of an HTCondor pool with GPU execution capabilities through OpenCL. Implementation is discussed from both the system setup and the software design standpoint. The GPU landscape is investigated and the efficiency of the system is evaluated via FastFourier Transform computations. Experimental results show that HTCondor GPU performance matches a dedicated GPU cluster.

AB - As the computational trend diverges from standard CPU computing, to encompass GPUs and other accelerators, the need to integrate these unused resources within existing systems becomes apparent. This paper presents the implementation of an HTCondor pool with GPU execution capabilities through OpenCL. Implementation is discussed from both the system setup and the software design standpoint. The GPU landscape is investigated and the efficiency of the system is evaluated via FastFourier Transform computations. Experimental results show that HTCondor GPU performance matches a dedicated GPU cluster.

KW - HPC

KW - HTCondor

KW - OpenCL

KW - GPU

M3 - Conference contribution

SP - 46

EP - 49

BT - Proceedings of the EMerging Technology (EMiT) Conference 2016

A2 - Rogers, B.D.

A2 - Topping, D.

A2 - Mantovani, F.

A2 - Bane, M.K.

PB - The Emerging Technology (EMiT) Conference

CY - Manchester

T2 - Emerging Technology Conference

Y2 - 2 June 2016 through 3 June 2016

ER -

Accelerating High-Throughput Computing through OpenCL

Abstract

Conference

Access to Document

Fingerprint

Cite this