Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster

Vincent W. Freeh, Feng Pan, Nandini Kappiah, David K Lowenthal, Rob Springer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

87 Citations (Scopus)

Abstract

Recently, energy has become an important issue in high-performance computing. For example, supercomputers that have energy in mind, such as BlueGene/L, have been built; the idea is to improve the energy efficiency of nodes. Our approach, which uses off-the-shelf, high-performance cluster nodes that are frequency scalable, allows energy saving by scaling down the CPU. This paper investigates the energy consumption and execution time of applications from a standard benchmark suite (NAS) on a power-scalable cluster. We study via direct measurement and simulation both intra-node and inter-node effects of memory and communication bottlenecks, respectively. Additionally, we compare energy consumption and execution time across different numbers of nodes. Our results show that a power-scalable cluster has the potential to save energy by scaling the processor down to lower energy levels. Furthermore, we found that for some programs, it is possible to both consume less energy and execute in less time when using a larger number of nodes, each at reduced energy. Additionally, we developed and validated a model that enables us to predict the energy-time tradeoff of larger clusters.

Original languageEnglish (US)
Title of host publicationProceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005
Volume2005
DOIs
StatePublished - 2005
Externally publishedYes
Event19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005 - Denver, CO, United States
Duration: Apr 4 2005Apr 8 2005

Other

Other19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005
CountryUnited States
CityDenver, CO
Period4/4/054/8/05

Fingerprint

Energy utilization
Supercomputers
Electron energy levels
Program processors
Energy efficiency
Energy conservation
Data storage equipment
Communication

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Freeh, V. W., Pan, F., Kappiah, N., Lowenthal, D. K., & Springer, R. (2005). Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. In Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005 (Vol. 2005). [1419817] https://doi.org/10.1109/IPDPS.2005.214

Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. / Freeh, Vincent W.; Pan, Feng; Kappiah, Nandini; Lowenthal, David K; Springer, Rob.

Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005. Vol. 2005 2005. 1419817.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Freeh, VW, Pan, F, Kappiah, N, Lowenthal, DK & Springer, R 2005, Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. in Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005. vol. 2005, 1419817, 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005, Denver, CO, United States, 4/4/05. https://doi.org/10.1109/IPDPS.2005.214
Freeh VW, Pan F, Kappiah N, Lowenthal DK, Springer R. Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. In Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005. Vol. 2005. 2005. 1419817 https://doi.org/10.1109/IPDPS.2005.214
Freeh, Vincent W. ; Pan, Feng ; Kappiah, Nandini ; Lowenthal, David K ; Springer, Rob. / Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005. Vol. 2005 2005.
@inproceedings{f8331a87ad60425f8b10410b4777880d,
title = "Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster",
abstract = "Recently, energy has become an important issue in high-performance computing. For example, supercomputers that have energy in mind, such as BlueGene/L, have been built; the idea is to improve the energy efficiency of nodes. Our approach, which uses off-the-shelf, high-performance cluster nodes that are frequency scalable, allows energy saving by scaling down the CPU. This paper investigates the energy consumption and execution time of applications from a standard benchmark suite (NAS) on a power-scalable cluster. We study via direct measurement and simulation both intra-node and inter-node effects of memory and communication bottlenecks, respectively. Additionally, we compare energy consumption and execution time across different numbers of nodes. Our results show that a power-scalable cluster has the potential to save energy by scaling the processor down to lower energy levels. Furthermore, we found that for some programs, it is possible to both consume less energy and execute in less time when using a larger number of nodes, each at reduced energy. Additionally, we developed and validated a model that enables us to predict the energy-time tradeoff of larger clusters.",
author = "Freeh, {Vincent W.} and Feng Pan and Nandini Kappiah and Lowenthal, {David K} and Rob Springer",
year = "2005",
doi = "10.1109/IPDPS.2005.214",
language = "English (US)",
isbn = "0769523129",
volume = "2005",
booktitle = "Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005",

}

TY - GEN

T1 - Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster

AU - Freeh, Vincent W.

AU - Pan, Feng

AU - Kappiah, Nandini

AU - Lowenthal, David K

AU - Springer, Rob

PY - 2005

Y1 - 2005

N2 - Recently, energy has become an important issue in high-performance computing. For example, supercomputers that have energy in mind, such as BlueGene/L, have been built; the idea is to improve the energy efficiency of nodes. Our approach, which uses off-the-shelf, high-performance cluster nodes that are frequency scalable, allows energy saving by scaling down the CPU. This paper investigates the energy consumption and execution time of applications from a standard benchmark suite (NAS) on a power-scalable cluster. We study via direct measurement and simulation both intra-node and inter-node effects of memory and communication bottlenecks, respectively. Additionally, we compare energy consumption and execution time across different numbers of nodes. Our results show that a power-scalable cluster has the potential to save energy by scaling the processor down to lower energy levels. Furthermore, we found that for some programs, it is possible to both consume less energy and execute in less time when using a larger number of nodes, each at reduced energy. Additionally, we developed and validated a model that enables us to predict the energy-time tradeoff of larger clusters.

AB - Recently, energy has become an important issue in high-performance computing. For example, supercomputers that have energy in mind, such as BlueGene/L, have been built; the idea is to improve the energy efficiency of nodes. Our approach, which uses off-the-shelf, high-performance cluster nodes that are frequency scalable, allows energy saving by scaling down the CPU. This paper investigates the energy consumption and execution time of applications from a standard benchmark suite (NAS) on a power-scalable cluster. We study via direct measurement and simulation both intra-node and inter-node effects of memory and communication bottlenecks, respectively. Additionally, we compare energy consumption and execution time across different numbers of nodes. Our results show that a power-scalable cluster has the potential to save energy by scaling the processor down to lower energy levels. Furthermore, we found that for some programs, it is possible to both consume less energy and execute in less time when using a larger number of nodes, each at reduced energy. Additionally, we developed and validated a model that enables us to predict the energy-time tradeoff of larger clusters.

UR - http://www.scopus.com/inward/record.url?scp=33746283629&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746283629&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2005.214

DO - 10.1109/IPDPS.2005.214

M3 - Conference contribution

SN - 0769523129

SN - 0769523129

SN - 9780769523125

VL - 2005

BT - Proceedings - 19th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2005

ER -