High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability

Ada Sedova; John D. Eblen; Reuben Budiardja; Arnold Tharrington; Jeremy C. Smith

doi:10.1109/P3HPC.2018.00004

High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability

Ada Sedova, John D. Eblen, Reuben Budiardja, Arnold Tharrington, Jeremy C. Smith

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

21 Scopus citations

Abstract

Highly-optimized parallel molecular dynamics programs have allowed researchers to achieve ground-breaking results in biological and materials sciences. This type of performance has come at the expense of portability: A significant effort is required for performance optimization on each new architecture. Using a metric that emphasizes speedup, we assess key accelerating programming components of four different best-performing molecular dynamics programs-GROMACS, NAMD, LAMMPS and CP2K-each having a particular scope of application, for contribution to performance and for portability. We use builds with and without these components, tested on HPC systems. We also analyze the code-bases to determine compliance with portability recommendations. We find that for all four programs, the contributions of the non-portable components to speed are essential to the programs' performances; without them we see a reduction in time-to-solution of a magnitude that is insufferable to domain scientists. This characterizes the performance efficiency that must be approached for good performance portability on a programmatic level, suggesting solutions to this difficult problem, which should come from developers, industry and funding institutions, and possibly new research in programming languages.

Original language	English
Title of host publication	Proceedings of P3HPC 2018
Subtitle of host publication	International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-13
Number of pages	13
ISBN (Electronic)	9781728102207
DOIs	https://doi.org/10.1109/P3HPC.2018.00004
State	Published - Jul 2 2018
Event	2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC, P3HPC 2018 - Dallas, United States Duration: Nov 16 2018 → …

Publication series

Name	Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC, P3HPC 2018
Country/Territory	United States
City	Dallas
Period	11/16/18 → …

Funding

This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/ downloads/doe-public-access-plan).

Keywords

CUDA-C
SIMD-instructions
heterogenous-architectures
molecular-dynamics
performance-portability

Access to Document

10.1109/P3HPC.2018.00004

Cite this

Sedova, A., Eblen, J. D., Budiardja, R., Tharrington, A., & Smith, J. C. (2018). High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability. In Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 1-13). Article 8639943 (Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/P3HPC.2018.00004

Sedova, Ada ; Eblen, John D. ; Budiardja, Reuben et al. / High-performance molecular dynamics simulation for biological and materials sciences : Challenges of performance portability. Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1-13 (Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{3140bf2408444abd95db7a36988b9ccd,

title = "High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability",

abstract = "Highly-optimized parallel molecular dynamics programs have allowed researchers to achieve ground-breaking results in biological and materials sciences. This type of performance has come at the expense of portability: A significant effort is required for performance optimization on each new architecture. Using a metric that emphasizes speedup, we assess key accelerating programming components of four different best-performing molecular dynamics programs-GROMACS, NAMD, LAMMPS and CP2K-each having a particular scope of application, for contribution to performance and for portability. We use builds with and without these components, tested on HPC systems. We also analyze the code-bases to determine compliance with portability recommendations. We find that for all four programs, the contributions of the non-portable components to speed are essential to the programs' performances; without them we see a reduction in time-to-solution of a magnitude that is insufferable to domain scientists. This characterizes the performance efficiency that must be approached for good performance portability on a programmatic level, suggesting solutions to this difficult problem, which should come from developers, industry and funding institutions, and possibly new research in programming languages.",

keywords = "CUDA-C, SIMD-instructions, heterogenous-architectures, molecular-dynamics, performance-portability",

author = "Ada Sedova and Eblen, {John D.} and Reuben Budiardja and Arnold Tharrington and Smith, {Jeremy C.}",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC, P3HPC 2018 ; Conference date: 16-11-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/P3HPC.2018.00004",

language = "English",

series = "Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1--13",

booktitle = "Proceedings of P3HPC 2018",

}

Sedova, A, Eblen, JD, Budiardja, R , Tharrington, A & Smith, JC 2018, High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability. in Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis., 8639943, Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 1-13, 2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC, P3HPC 2018, Dallas, United States, 11/16/18. https://doi.org/10.1109/P3HPC.2018.00004

High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability. / Sedova, Ada; Eblen, John D.; Budiardja, Reuben et al.
Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1-13 8639943 (Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis).

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

TY - GEN

T1 - High-performance molecular dynamics simulation for biological and materials sciences

T2 - 2018 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC, P3HPC 2018

AU - Sedova, Ada

AU - Eblen, John D.

AU - Budiardja, Reuben

AU - Tharrington, Arnold

AU - Smith, Jeremy C.

PY - 2018/7/2

Y1 - 2018/7/2

N2 - Highly-optimized parallel molecular dynamics programs have allowed researchers to achieve ground-breaking results in biological and materials sciences. This type of performance has come at the expense of portability: A significant effort is required for performance optimization on each new architecture. Using a metric that emphasizes speedup, we assess key accelerating programming components of four different best-performing molecular dynamics programs-GROMACS, NAMD, LAMMPS and CP2K-each having a particular scope of application, for contribution to performance and for portability. We use builds with and without these components, tested on HPC systems. We also analyze the code-bases to determine compliance with portability recommendations. We find that for all four programs, the contributions of the non-portable components to speed are essential to the programs' performances; without them we see a reduction in time-to-solution of a magnitude that is insufferable to domain scientists. This characterizes the performance efficiency that must be approached for good performance portability on a programmatic level, suggesting solutions to this difficult problem, which should come from developers, industry and funding institutions, and possibly new research in programming languages.

AB - Highly-optimized parallel molecular dynamics programs have allowed researchers to achieve ground-breaking results in biological and materials sciences. This type of performance has come at the expense of portability: A significant effort is required for performance optimization on each new architecture. Using a metric that emphasizes speedup, we assess key accelerating programming components of four different best-performing molecular dynamics programs-GROMACS, NAMD, LAMMPS and CP2K-each having a particular scope of application, for contribution to performance and for portability. We use builds with and without these components, tested on HPC systems. We also analyze the code-bases to determine compliance with portability recommendations. We find that for all four programs, the contributions of the non-portable components to speed are essential to the programs' performances; without them we see a reduction in time-to-solution of a magnitude that is insufferable to domain scientists. This characterizes the performance efficiency that must be approached for good performance portability on a programmatic level, suggesting solutions to this difficult problem, which should come from developers, industry and funding institutions, and possibly new research in programming languages.

KW - CUDA-C

KW - SIMD-instructions

KW - heterogenous-architectures

KW - molecular-dynamics

KW - performance-portability

UR - http://www.scopus.com/inward/record.url?scp=85063105896&partnerID=8YFLogxK

U2 - 10.1109/P3HPC.2018.00004

DO - 10.1109/P3HPC.2018.00004

M3 - Conference contribution

AN - SCOPUS:85063105896

T3 - Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 1

EP - 13

BT - Proceedings of P3HPC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 16 November 2018

ER -

Sedova A, Eblen JD, Budiardja R , Tharrington A, Smith JC. High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability. In Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1-13. 8639943. (Proceedings of P3HPC 2018: International Workshop on Performance, Portability and Productivity in HPC, Held in conjunction with SC 2018: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/P3HPC.2018.00004

High-performance molecular dynamics simulation for biological and materials sciences: Challenges of performance portability

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this