Application configuration selection for energy-efficient execution on multicore systems
Abstract
Balanced performance and energy consumption are incorporated in the design of modern computer systems. Several runtime factors, such as concurrency levels, thread mapping strategies, and dynamic voltage and frequency scaling (DVFS) should be considered in order to achieve optimal energy efficiency fora workload. Selecting appropriate run-time factors, however, is one of the most challenging tasks because the run-time factors are architecture-specific and workload-specific. And while most existing works concentrate on either static analysis of the workload or run-time prediction results, we present a hybrid two-step method that utilizes concurrency levels and DVFS settings to achieve the energy efficiency configuration for a worldoad. The experimental results based on a Xeon E5620 server with NPB and PARSEC benchmark suites show that the model is able to predict the energy efficient configuration accurately. On average, an additional 10% EDP (Energy Delay Product) saving is obtained by using run-time DVFS for the entire system. An off-line optimal solution is used to compare with the proposed scheme. Finally, the experimental results show that the average extra EDP saved by the optimal solution is within 5% on selective parallel benchmarks.
- Authors:
-
- Wayne State Univ., Detroit, MI (United States)
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Publication Date:
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
- Sponsoring Org.:
- USDOE Office of Science (SC); National Science Foundation (NSF)
- OSTI Identifier:
- 1261493
- Alternate Identifier(s):
- OSTI ID: 1359446
- Grant/Contract Number:
- AC05-00OR22725; CNS-1205338
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Journal of Parallel and Distributed Computing
- Additional Journal Information:
- Journal Volume: 87; Journal Issue: C; Journal ID: ISSN 0743-7315
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Energy consumption; High performance computing; Speedup model; Power model; Parallel; POWER MANAGEMENT; PERFORMANCE
Citation Formats
Wang, Shinan, Luo, Bing, Shi, Weisong, and Tiwari, Devesh. Application configuration selection for energy-efficient execution on multicore systems. United States: N. p., 2015.
Web. doi:10.1016/j.jpdc.2015.09.003.
Wang, Shinan, Luo, Bing, Shi, Weisong, & Tiwari, Devesh. Application configuration selection for energy-efficient execution on multicore systems. United States. https://doi.org/10.1016/j.jpdc.2015.09.003
Wang, Shinan, Luo, Bing, Shi, Weisong, and Tiwari, Devesh. Mon .
"Application configuration selection for energy-efficient execution on multicore systems". United States. https://doi.org/10.1016/j.jpdc.2015.09.003. https://www.osti.gov/servlets/purl/1261493.
@article{osti_1261493,
title = {Application configuration selection for energy-efficient execution on multicore systems},
author = {Wang, Shinan and Luo, Bing and Shi, Weisong and Tiwari, Devesh},
abstractNote = {Balanced performance and energy consumption are incorporated in the design of modern computer systems. Several runtime factors, such as concurrency levels, thread mapping strategies, and dynamic voltage and frequency scaling (DVFS) should be considered in order to achieve optimal energy efficiency fora workload. Selecting appropriate run-time factors, however, is one of the most challenging tasks because the run-time factors are architecture-specific and workload-specific. And while most existing works concentrate on either static analysis of the workload or run-time prediction results, we present a hybrid two-step method that utilizes concurrency levels and DVFS settings to achieve the energy efficiency configuration for a worldoad. The experimental results based on a Xeon E5620 server with NPB and PARSEC benchmark suites show that the model is able to predict the energy efficient configuration accurately. On average, an additional 10% EDP (Energy Delay Product) saving is obtained by using run-time DVFS for the entire system. An off-line optimal solution is used to compare with the proposed scheme. Finally, the experimental results show that the average extra EDP saved by the optimal solution is within 5% on selective parallel benchmarks.},
doi = {10.1016/j.jpdc.2015.09.003},
journal = {Journal of Parallel and Distributed Computing},
number = C,
volume = 87,
place = {United States},
year = {Mon Sep 21 00:00:00 EDT 2015},
month = {Mon Sep 21 00:00:00 EDT 2015}
}
Web of Science
Works referenced in this record:
The effect of state-saving in optimistic simulation on a cache-coherent non-uniform memory access architecture
conference, January 1999
- Carothers, Christopher D.; Perumalla, Kalyan S.; Fujimoto, Richard M.
- Proceedings of the 31st conference on Winter simulation Simulation---a bridge to the future - WSC '99
A machine learning-based approach for thread mapping on transactional memory applications
conference, December 2011
- Castro, Marcio; Goes, Luis Fabricio Wanderley; Ribeiro, Christiane Pousa
- 2011 18th International Conference on High Performance Computing (HiPC)
Power challenges may end the multicore era
journal, February 2013
- Esmaeilzadeh, Hadi; Blem, Emily; Amant, Renée St.
- Communications of the ACM, Vol. 56, Issue 2
Power-Aware Speedup
conference, March 2007
- Ge, Rong; Cameron, Kirk W.
- 2007 IEEE International Parallel and Distributed Processing Symposium
Modeling and evaluating energy-performance efficiency of parallel processing on multicore based power aware systems
conference, May 2009
- Ge, Rong; Feng, Xizhou; Cameron, Kirk W.
- Distributed Processing (IPDPS), 2009 IEEE International Symposium on Parallel & Distributed Processing
Portable, scalable, per-core power estimation for intelligent resource management
conference, August 2010
- Goel, Bhavishya; McKee, Sally A.; Gioiosa, Roberto
- 2010 International Conference on Green Computing (Green Comp)
Process variation aware thread mapping for Chip Multiprocessors
conference, April 2009
- Hong, S.; Narayanan, S. H. K.; Kandemir, M.
- 2009 Design, Automation & Test in Europe Conference & Exhibition (DATE'09)
Live, Runtime Phase Monitoring and Prediction on Real Systems with Application to Dynamic Power Management
conference, December 2006
- Isci, Canturk; Contreras, Gilberto; Martonosi, Margaret
- 2006 39th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'06)
Predicting Potential Speedup of Serial Code via Lightweight Profiling and Emulations with Memory Performance Model
conference, May 2012
- Kim, Minjang; Kumar, Pranith; Kim, Hyesoon
- 2012 IEEE International Symposium on Parallel & Distributed Processing (IPDPS), 2012 IEEE 26th International Parallel and Distributed Processing Symposium
Run-time modeling and estimation of operating system power consumption
journal, June 2003
- Li, Tao; John, Lizy Kurian
- ACM SIGMETRICS Performance Evaluation Review, Vol. 31, Issue 1
SolarCore: Solar energy driven multi-core architecture power management
conference, February 2011
- Li, Chao; Zhang, Wangyuan; Cho, Chang-Burm
- 2011 IEEE 17th International Symposium on High Performance Computer Architecture (HPCA)
Green governors: A framework for Continuously Adaptive DVFS
conference, July 2011
- Spiliopoulos, Vasileios; Kaxiras, Stefanos; Keramidas, Georgios
- 2011 International Green Computing Conference (IGCC), 2011 International Green Computing Conference and Workshops
SPAN: A software power analyzer for multicore computer systems
journal, March 2011
- Wang, Shinan; Chen, Hui; Shi, Weisong
- Sustainable Computing: Informatics and Systems, Vol. 1, Issue 1
Extending Amdahl's law in the multicore era
journal, October 2009
- Yao, Erlin; Bao, Yungang; Tan, Guangming
- ACM SIGMETRICS Performance Evaluation Review, Vol. 37, Issue 2
Works referencing / citing this record:
Dynamic Voltage Frequency Scaling Simulator for Real Workflows Energy-Aware Management in Green Cloud Computing
journal, January 2017
- Cotes-Ruiz, Iván Tomás; Prado, Rocío P.; García-Galán, Sebastián
- PLOS ONE, Vol. 12, Issue 1
Dynamic Voltage Frequency Scaling Simulator for Real Workflows Energy-Aware Management in Green Cloud Computing
journal, January 2017
- Cotes-Ruiz, Iván Tomás; Prado, Rocío P.; García-Galán, Sebastián
- PLOS ONE, Vol. 12, Issue 1