Energy-aware metrics for benchmarking heterogeneous systems

SN McIntosh-Smith; T Wilson; RB Sessions; A.A. Ibarra; JC Crisp

Energy-aware metrics for benchmarking heterogeneous systems

SN McIntosh-Smith, T Wilson, RB Sessions, A.A. Ibarra, JC Crisp

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

Abstract

With the advent of heterogeneous computing systems con- sisting of multi-core CPUs and many-core GPUs, robust methods are needed to facilitate fair benchmark compar- isons between different systems. In this paper we present a benchmarking methodology for measuring a number of per- formance metrics for heterogeneous systems. Methods for comparing performance and energy efficiency are included. Consideration is given to further metrics, such as associ- ated runnings costs and even carbon emissions. We give a case study for these metrics applied to BUDE, a molecular mechanics-based docking application that has been ported to OpenCL at the University of Bristol.

Translated title of the contribution	Energy-aware metrics for benchmarking heterogeneous systems
Original language	English
Title of host publication	ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10)
Publication status	Published - 15 Nov 2010

Bibliographical note

Name and Venue of Event: ACM / IEEE Supercomputing 2010 (SC10), New Orleans, November 13-19, 2010
Conference Organiser: ACM / IEEE

Access to Document

http://www.dcs.warwick.ac.uk/~sdh/pmbs10/pmbs10/Workshop_Programme_files/energyaware.pdf

Fingerprint Dive into the research topics of 'Energy-aware metrics for benchmarking heterogeneous systems'. Together they form a unique fingerprint.

HPC (High Performance Computing) facility
Caroline E Gardiner (Manager)
IT Services
Facility/equipment: Facility

Cite this

McIntosh-Smith, SN., Wilson, T., Sessions, RB., Ibarra, A. A., & Crisp, JC. (2010). Energy-aware metrics for benchmarking heterogeneous systems. In ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10) http://www.dcs.warwick.ac.uk/~sdh/pmbs10/pmbs10/Workshop_Programme_files/energyaware.pdf

@inproceedings{05fe865af04e48e49d55ee02f5db1794,

title = "Energy-aware metrics for benchmarking heterogeneous systems",

abstract = "With the advent of heterogeneous computing systems con- sisting of multi-core CPUs and many-core GPUs, robust methods are needed to facilitate fair benchmark compar- isons between different systems. In this paper we present a benchmarking methodology for measuring a number of per- formance metrics for heterogeneous systems. Methods for comparing performance and energy efficiency are included. Consideration is given to further metrics, such as associ- ated runnings costs and even carbon emissions. We give a case study for these metrics applied to BUDE, a molecular mechanics-based docking application that has been ported to OpenCL at the University of Bristol.",

author = "SN McIntosh-Smith and T Wilson and RB Sessions and A.A. Ibarra and JC Crisp",

note = "Name and Venue of Event: ACM / IEEE Supercomputing 2010 (SC10), New Orleans, November 13-19, 2010 Conference Organiser: ACM / IEEE",

year = "2010",

month = nov,

day = "15",

language = "English",

booktitle = "ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10)",

}

McIntosh-Smith, SN, Wilson, T, Sessions, RB, Ibarra, AA & Crisp, JC 2010, Energy-aware metrics for benchmarking heterogeneous systems. in ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10). <http://www.dcs.warwick.ac.uk/~sdh/pmbs10/pmbs10/Workshop_Programme_files/energyaware.pdf>

Energy-aware metrics for benchmarking heterogeneous systems. / McIntosh-Smith, SN; Wilson, T; Sessions, RB; Ibarra, A.A.; Crisp, JC.

ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10). 2010.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Energy-aware metrics for benchmarking heterogeneous systems

AU - McIntosh-Smith, SN

AU - Wilson, T

AU - Sessions, RB

AU - Ibarra, A.A.

AU - Crisp, JC

N1 - Name and Venue of Event: ACM / IEEE Supercomputing 2010 (SC10), New Orleans, November 13-19, 2010 Conference Organiser: ACM / IEEE

PY - 2010/11/15

Y1 - 2010/11/15

N2 - With the advent of heterogeneous computing systems con- sisting of multi-core CPUs and many-core GPUs, robust methods are needed to facilitate fair benchmark compar- isons between different systems. In this paper we present a benchmarking methodology for measuring a number of per- formance metrics for heterogeneous systems. Methods for comparing performance and energy efficiency are included. Consideration is given to further metrics, such as associ- ated runnings costs and even carbon emissions. We give a case study for these metrics applied to BUDE, a molecular mechanics-based docking application that has been ported to OpenCL at the University of Bristol.

AB - With the advent of heterogeneous computing systems con- sisting of multi-core CPUs and many-core GPUs, robust methods are needed to facilitate fair benchmark compar- isons between different systems. In this paper we present a benchmarking methodology for measuring a number of per- formance metrics for heterogeneous systems. Methods for comparing performance and energy efficiency are included. Consideration is given to further metrics, such as associ- ated runnings costs and even carbon emissions. We give a case study for these metrics applied to BUDE, a molecular mechanics-based docking application that has been ported to OpenCL at the University of Bristol.

M3 - Conference Contribution (Conference Proceeding)

BT - ACM SIGMETRICS 1st International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 10)

ER -

Energy-aware metrics for benchmarking heterogeneous systems

Abstract

Bibliographical note

Access to Document

HPC (High Performance Computing) facility

Cite this