BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics

Edd Bennett; Biagio Lucini; Luigi Del Debbio; Kirk Jordan; Agostino Patella; Claudio Pica

doi:10.1109/HPCSim.2016.7568421

BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics

Edd Bennett, Biagio Lucini, Luigi Del Debbio, Kirk Jordan, Agostino Patella, Claudio Pica

The University of Edinburgh

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

Abstract

Lattice Quantum ChromoDynamics (QCD), and by extension its parent field, Lattice Gauge Theory (LGT), make up a significant fraction of supercomputing cycles worldwide. As such, it would be irresponsible not to evaluate machines' suitability for such applications. To this end, a benchmark has been developed to assess the performance of LGT applications on modern HPC platforms. Distinct from previous QCD-based benchmarks, this allows probing the behaviour of a variety of theories, which allows varying the ratio of demands between on-node computations and inter-node communications. The results of testing this benchmark on various recent HPC platforms are presented, and directions for future development are discussed.

Original language	English
Title of host publication	Proceedings of the 14th International Conference on High Performance Computing & Simulation
Editors	Vesna Zeljkovic, Waleed W. Smari
Publisher	IEEE
Publication date	13. Sept 2016
Pages	834-839
ISBN (Print)	978-1-5090-2089-8
ISBN (Electronic)	978-1-5090-2088-1
DOIs	https://doi.org/10.1109/HPCSim.2016.7568421
Publication status	Published - 13. Sept 2016
Event	14th International Conference on High Performance Computing & Simulation - Innsbruck, Austria Duration: 18. Jul 2016 → 22. Jul 2018 Conference number: 14

Conference

Conference	14th International Conference on High Performance Computing & Simulation
Number	14
Country/Territory	Austria
City	Innsbruck
Period	18/07/2016 → 22/07/2018

Keywords

Xeon Phi
benchmarking
beyond the standard model
quantum chromodynamics

Documents & Links

10.1109/HPCSim.2016.7568421

Cite this

@inproceedings{ba3ea62aa5f44241900a75b910d21911,

title = "BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics",

abstract = "Lattice Quantum ChromoDynamics (QCD), and by extension its parent field, Lattice Gauge Theory (LGT), make up a significant fraction of supercomputing cycles worldwide. As such, it would be irresponsible not to evaluate machines' suitability for such applications. To this end, a benchmark has been developed to assess the performance of LGT applications on modern HPC platforms. Distinct from previous QCD-based benchmarks, this allows probing the behaviour of a variety of theories, which allows varying the ratio of demands between on-node computations and inter-node communications. The results of testing this benchmark on various recent HPC platforms are presented, and directions for future development are discussed.",

keywords = "Xeon Phi, benchmarking, beyond the standard model, quantum chromodynamics",

author = "Edd Bennett and Biagio Lucini and {Del Debbio}, Luigi and Kirk Jordan and Agostino Patella and Claudio Pica",

year = "2016",

month = sep,

day = "13",

doi = "10.1109/HPCSim.2016.7568421",

language = "English",

isbn = "978-1-5090-2089-8",

pages = "834--839",

editor = "Vesna Zeljkovic and Smari, {Waleed W.}",

booktitle = "Proceedings of the 14th International Conference on High Performance Computing & Simulation",

publisher = "IEEE",

address = "United States",

note = "14th International Conference on High Performance Computing & Simulation ; Conference date: 18-07-2016 Through 22-07-2018",

}

Bennett, E, Lucini, B, Del Debbio, L, Jordan, K, Patella, A & Pica, C 2016, BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics. in V Zeljkovic & WW Smari (eds), Proceedings of the 14th International Conference on High Performance Computing & Simulation. IEEE, pp. 834-839, 14th International Conference on High Performance Computing & Simulation, Innsbruck, Austria, 18/07/2016. https://doi.org/10.1109/HPCSim.2016.7568421

BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics. / Bennett, Edd; Lucini, Biagio; Del Debbio, Luigi et al.
Proceedings of the 14th International Conference on High Performance Computing & Simulation. ed. / Vesna Zeljkovic; Waleed W. Smari. IEEE, 2016. p. 834-839.

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

TY - GEN

T1 - BSMBench

T2 - 14th International Conference on High Performance Computing & Simulation

AU - Bennett, Edd

AU - Lucini, Biagio

AU - Del Debbio, Luigi

AU - Jordan, Kirk

AU - Patella, Agostino

AU - Pica, Claudio

N1 - Conference code: 14

PY - 2016/9/13

Y1 - 2016/9/13

N2 - Lattice Quantum ChromoDynamics (QCD), and by extension its parent field, Lattice Gauge Theory (LGT), make up a significant fraction of supercomputing cycles worldwide. As such, it would be irresponsible not to evaluate machines' suitability for such applications. To this end, a benchmark has been developed to assess the performance of LGT applications on modern HPC platforms. Distinct from previous QCD-based benchmarks, this allows probing the behaviour of a variety of theories, which allows varying the ratio of demands between on-node computations and inter-node communications. The results of testing this benchmark on various recent HPC platforms are presented, and directions for future development are discussed.

AB - Lattice Quantum ChromoDynamics (QCD), and by extension its parent field, Lattice Gauge Theory (LGT), make up a significant fraction of supercomputing cycles worldwide. As such, it would be irresponsible not to evaluate machines' suitability for such applications. To this end, a benchmark has been developed to assess the performance of LGT applications on modern HPC platforms. Distinct from previous QCD-based benchmarks, this allows probing the behaviour of a variety of theories, which allows varying the ratio of demands between on-node computations and inter-node communications. The results of testing this benchmark on various recent HPC platforms are presented, and directions for future development are discussed.

KW - Xeon Phi

KW - benchmarking

KW - beyond the standard model

KW - quantum chromodynamics

U2 - 10.1109/HPCSim.2016.7568421

DO - 10.1109/HPCSim.2016.7568421

M3 - Article in proceedings

SN - 978-1-5090-2089-8

SP - 834

EP - 839

BT - Proceedings of the 14th International Conference on High Performance Computing & Simulation

A2 - Zeljkovic, Vesna

A2 - Smari, Waleed W.

PB - IEEE

Y2 - 18 July 2016 through 22 July 2018

ER -

BSMBench: A flexible and scalable HPC benchmark from beyond the standard model physics

Abstract

Conference

Keywords

Documents & Links

Fingerprint

Cite this