On the energy footprint of I/O management in Exascale HPC systems
Abstract
The advent of unprecedentedly scalable yet energy hungry Exascale supercomputers poses a major challenge in sustaining a high performance-per-watt ratio. With I/O management acquiring a crucial role in supporting scientific simulations, various I/O management approaches have been proposed to achieve high performance and scalability. But, the details of how these approaches affect energy consumption have not been studied yet. Therefore, this paper aims to explore how much energy a supercomputer consumes while running scientific simulations when adopting various I/O management approaches. In particular, we closely examine three radically different I/O schemes including time partitioning, dedicated cores, and dedicated nodes. To accomplish this, we implement the three approaches within the Damaris I/O middleware and perform extensive experiments with one of the target HPC applications of the Blue Waters sustained-petaflop supercomputer project: the CM1 atmospheric model. Our experimental results obtained on the French Grid'5000 platform highlight the differences among these three approaches and illustrate in which way various configurations of the application and of the system can impact performance and energy consumption. Moreover, we propose and validate a mathematical model that estimates the energy consumption of a HPC simulation under different I/O approaches. This proposed model gives hints to pre-select the mostmore »
- Authors:
-
- Ecole Normale Supérieure of Rennes (ENS) (France); Argonne National Lab. (ANL), Argonne, IL (United States)
- Inra, Rennes (France)
- Centre National de la Recherche Scientifique (CNRS), Annecy-le-Vieux (France). Research Inst. of Computer Science and Random Systems (IRISA)
- Publication Date:
- Research Org.:
- Argonne National Laboratory (ANL), Argonne, IL (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- OSTI Identifier:
- 1390829
- Grant/Contract Number:
- AC02-06CH11357
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Future Generations Computer Systems
- Additional Journal Information:
- Journal Volume: 62; Journal Issue: C; Journal ID: ISSN 0167-739X
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; I/O; Damaris; Dedicated Cores; Dedicated Nodes; Energy; Exascale Computing
Citation Formats
Dorier, Matthieu, Yildiz, Orcun, Ibrahim, Shadi, Orgerie, Anne-Cécile, and Antoniu, Gabriel. On the energy footprint of I/O management in Exascale HPC systems. United States: N. p., 2016.
Web. doi:10.1016/j.future.2016.03.002.
Dorier, Matthieu, Yildiz, Orcun, Ibrahim, Shadi, Orgerie, Anne-Cécile, & Antoniu, Gabriel. On the energy footprint of I/O management in Exascale HPC systems. United States. https://doi.org/10.1016/j.future.2016.03.002
Dorier, Matthieu, Yildiz, Orcun, Ibrahim, Shadi, Orgerie, Anne-Cécile, and Antoniu, Gabriel. Mon .
"On the energy footprint of I/O management in Exascale HPC systems". United States. https://doi.org/10.1016/j.future.2016.03.002. https://www.osti.gov/servlets/purl/1390829.
@article{osti_1390829,
title = {On the energy footprint of I/O management in Exascale HPC systems},
author = {Dorier, Matthieu and Yildiz, Orcun and Ibrahim, Shadi and Orgerie, Anne-Cécile and Antoniu, Gabriel},
abstractNote = {The advent of unprecedentedly scalable yet energy hungry Exascale supercomputers poses a major challenge in sustaining a high performance-per-watt ratio. With I/O management acquiring a crucial role in supporting scientific simulations, various I/O management approaches have been proposed to achieve high performance and scalability. But, the details of how these approaches affect energy consumption have not been studied yet. Therefore, this paper aims to explore how much energy a supercomputer consumes while running scientific simulations when adopting various I/O management approaches. In particular, we closely examine three radically different I/O schemes including time partitioning, dedicated cores, and dedicated nodes. To accomplish this, we implement the three approaches within the Damaris I/O middleware and perform extensive experiments with one of the target HPC applications of the Blue Waters sustained-petaflop supercomputer project: the CM1 atmospheric model. Our experimental results obtained on the French Grid'5000 platform highlight the differences among these three approaches and illustrate in which way various configurations of the application and of the system can impact performance and energy consumption. Moreover, we propose and validate a mathematical model that estimates the energy consumption of a HPC simulation under different I/O approaches. This proposed model gives hints to pre-select the most energy-efficient I/O approach for a particular simulation on a particular HPC system and therefore provides a step towards energy-efficient HPC simulations in Exascale systems. To the best of our knowledge, our work provides the first in-depth look into the energy-performance tradeoffs of I/O management approaches.},
doi = {10.1016/j.future.2016.03.002},
journal = {Future Generations Computer Systems},
number = C,
volume = 62,
place = {United States},
year = {Mon Mar 21 00:00:00 EDT 2016},
month = {Mon Mar 21 00:00:00 EDT 2016}
}
Web of Science
Works referenced in this record:
Enabling high-speed asynchronous data extraction and transfer using DART
journal, January 2010
- Docan, Ciprian; Parashar, Manish; Klasky, Scott
- Concurrency and Computation: Practice and Experience
Damaris/Viz: A nonintrusive, adaptable and user-friendly in situ visualization framework
conference, October 2013
- Dorier, Matthieu; Sisneros, Robert; Peterka, Tom
- 2013 IEEE Symposium on Large-Scale Data Analysis and Visualization (LDAV)
A performance and energy analysis of I/O management approaches for exascale systems
conference, June 2014
- Yildiz, Orcun; Dorier, Matthieu; Ibrahim, Shadi
- Proceedings of the sixth international workshop on Data intensive distributed computing
Evaluation of Collective I/O Implementations on Parallel Architectures
journal, August 2001
- Dickens, Phillip M.; Thakur, Rajeev
- Journal of Parallel and Distributed Computing, Vol. 61, Issue 8
High-level buffering for hiding periodic output cost in scientific simulations
journal, March 2006
- Ma, X.; Lee, J.; Winslett, M.
- IEEE Transactions on Parallel and Distributed Systems, Vol. 17, Issue 3
A Benchmark Simulation for Moist Nonhydrostatic Numerical Models
journal, December 2002
- Bryan, George H.; Fritsch, J. Michael
- Monthly Weather Review, Vol. 130, Issue 12
A survey on techniques for improving the energy efficiency of large-scale distributed systems
journal, April 2014
- Orgerie, Anne-Cecile; Assuncao, Marcos Dias de; Lefevre, Laurent
- ACM Computing Surveys, Vol. 46, Issue 4
Governing energy consumption in Hadoop through CPU frequency scaling: An analysis
journal, January 2016
- Ibrahim, Shadi; Phan, Tien-Dat; Carpen-Amarie, Alexandra
- Future Generation Computer Systems, Vol. 54
ACPI thermal sensing and control in the PC
conference, September 1998
- Steele, J.
- Wescon/98. Conference Proceedings (Cat. No.98CH36265)
Energy efficiency in high-performance computing with and without knowledge of applications and services
journal, July 2013
- Diouri, Mohammed EM; Chetsa, Ghislain L. Tsafack; Glück, Olivier
- The International Journal of High Performance Computing Applications, Vol. 27, Issue 3