Green computing on graphics processing units

Author:

Magoulès Frédéric1,Ahamed Abal‐Kassim Cheik1,Suzuki Atsushi1

Affiliation:

1. CentraleSupélec Grande Voie des Vignes 92295 Châtenay‐Malabry Cedex France

Abstract

SummaryTo answer the question ‘How much energy is consumed for a numerical simulation running on Graphic Processing Unit?’, an experimental protocol is here established. The current provided to a graphic processing unit (GPU) during computation is directly measured using amperometric clamps. Signal processing on the intensity of the current of the power supplied to a GPU, with noise reduction technique, gives precise timing of GPU states, which allow establishing an energy consumption model of the GPU. Energy consumption of each operation, memory copy, vector addition, and element wise product is precisely measured to tune and validate the energy consumption model. The accuracy of the proposed energy consumption model compared to measurements is finally illustrated on a conjugate gradient method for a problem discretized by a finite element method. Copyright © 2015 John Wiley & Sons, Ltd.

Funder

CUDA Research Center at Ecole Centrale Paris (France)

Publisher

Wiley

Reference37 articles.

1. The LINPACK Benchmark: past, present and future

2. LiN SuchomelB Osei‐KuffuorD LiR SaadY.ITSOL 2010. (Available from:www-users.cs.umn.edu/~saad/software/ITSOL/index.html) (Accessed on 22 September 2015).

3. BalayS AdamsMF BrownJ BruneP BuschelmanK EijkhoutV GroppWD KaushikD KnepleyMG McInnesLC RuppK SmithBF ZhangH.PETSc users manual.Technical Report ANL‐95/11 ‐ Revision 3.4 Argonne National Laboratory:Argonne IL USA 2013.

4. LiuW DuZ HiaoY DavidAB XuC.A waterfall model to achieve energy efficient tasks mapping for large scale GPU clusters.Proceedings of the IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum Shanghai China 16‐20 may 2011 IEEE 2011;82–92.

5. HuoH ShengC HuX WuB.An energy efficient task scheduling scheme for heterogeneous GPU‐enhanced clusters.Proceedings of the International Conference on Systems and Informatics Yantai China May 19‐20 2012 IEEE 2012;623–627.

Cited by 10 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Point-Cloud-based Deep Learning Models for Finite Element Analysis;2022 21st International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES);2022-10

2. Multilayer Perceptron-based Surrogate Models for Finite Element Analysis;2022 21st International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES);2022-10

3. Competitiveness of a Non-Linear Block-Space GPU Thread Map for Simplex Domains;IEEE Transactions on Parallel and Distributed Systems;2018-12-01

4. Convergence Detection of Asynchronous Iterations Based on Modified Recursive Doubling;2018 17th International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES);2018-10

5. Asynchronous Iterations of Parareal Algorithm for Option Pricing Models;Mathematics;2018-03-21

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.7亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2025 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3