计算机科学 ›› 2015, Vol. 42 ›› Issue (2): 18-23.doi: 10.11896/j.issn.1002-137X.2015.02.004
王科特,王力生,廖新考
WANG Ke-te, WANG Li-sheng and LIAO Xin-kao
摘要: 针对具有独立DVFS的多核处理器系统,提出了一种K线程低能耗模型的并行任务调度优化算法(Tasks Optimization based on Energy-Effectiveness Model,TO-EEM)。与传统的并行任务节能调度相比,该算法的主要目标是不仅通过降低处理器频率来减少处理器瞬时功耗,而且结合并行任务间的同步互斥所造成的线程阻塞情况,合理分配线程资源来减少线程同步时间,优化并行性能;保证任务在一定的并行加速比性能前提下,提高资源利用率,减少能耗,达到程序能耗和性能之间的折衷。文中进行了大量模拟实验,结果证明提出的任务优化模型算法节能效果明显,能有效降低处理器的功耗,并始终保持线性加速比。
[1] 李新,贾智平,鞠雷,等.一种面向同构集群系统的并行任务节能调度优化方法[J].计算机学报,2012,35(3):591-602 [2] 张东松,吴飞,陈芳园,等.开销敏感的多处理器最优节能实时调度算法[J].计算机学报,2012,35(6):1297-1312 [3] 林宇晗,孔繁鑫,徐惠婷,等.线性加速比并行实时任务的节能研究[J].计算机学报,2013,36(2):384-393 [4] 张东松,吴彤,陈芳园,等.多核系统中基于Global+EDF的在线节能实时调度算法[J].软件学报,2012,23(4):996-1009 [5] 徐卫志,宋风龙,刘志勇,等.众核处理器片上同步机制和评估方法研究[J].计算机学报,2010,33(10):1777-1787 [6] 王之元.产出率并行加速比模型[J].计算机工程,2011,37(5):10-12 [7] BansaI N,Kimbrel T,Prush K.Speed sca1ing to manage energy and temperature[J].Journal of the ACM,2007,4(1):1-39 [8] Wierman A,Andrew L L H,Tang Ao.Power-aware speed sca-ling in processor sharing systems:Optimality and robustness[J].Performance Evaluation,2012(69):601-622 [9] Burd T D,Brodersen R W.Energy efficient cmos microprocessor design[C]∥Proc.of The HICSS Conference.Jan.1995 [10] Benoit A,Cole M,Gilmore S,et al.Flexible Skeletal Programming with eSkel[J].Submitted to EuroPar 2005.Lisbon,Portagal,2005,3648:761-770 [11] MacDonald S,Anvik J,Bromling S,et al.From patterns toframeworks to parallel programs[J].Parallel Computing,2002(28):1663-1683 [12] Bromling S,MacDonald S,Anvik J,et al.Pattern-based Parallel Programming[C]∥Proceedings of the International Conference on Parallel Processing (ICPP’02).2002 [13] Macdonald S.Deferring Design Pattern Decisions and Automa-ting Structural Pattern Changes using a Design-Pattern-Based Programming System[J].ACM Transactions on Programming Languages and Systems,2007:1-45 [14] Yzelman A N,Bisseling R H.An object-oriented bulk synchronous parallel library for multicore programming[M]∥Concurrency and Computation:Practice and Experience Concurrency Computat.:Pract.Exper.2012,24(5):533-553 [15] Yu Ce,Xu Zhen.ParaModel:A Visual Modeling and Code Skeleton Generation System for Programming Parallel Applications[J].ACM SIGPLAN Notices,2008,43(4) [16] Benoit A,Cole M.Two Fundamental Concepts in Skeletal Parallel Programming[C]∥ICCS 2005,LNCS 3515.2005:764-771 [17] Matsuzaki K,Emoto K.Lessons from implementing the biCG-Stab method with SkeTo library[C]∥International Conference on Functional Programming.2010:15-24 [18] Benoit A,Cole M,Hillston J,et al.Using eSkel to implement the multiple baseline stereo application.http://homepages.inf.ac.uk.mic.pubs/parco05.pdf [19] Siu S,Singh A.Design patterns for parallel computing using a network of processors[C]∥Sixth IEEE International Symposium on High Performance Distributed Computing.Oregon,August 1997:293-304 [20] Romero P,Cox R,du Boulay B,et al.A survey of external representations employed in object-oriented programming environments[J].Journal of visual languages and Computing,2003(14):387-419 [21] Vanneschi M.The programming model of ASSIST,an environment for parallel and distributed portable applications[J].Parallel Computing,2002,28 (12):1709-1732 [22] Aldinucci M,Coppola M,Danelutto M,et al.High level grid programming with ASSIST[J].Computational Methods in Science and Technology,2006,12(1):21-32 [23] Sun Jun-qing,Semiconductor M,Clara S.An Effective Execution Time Approximation Method for Parallel Computing[J].IEEE Transactions on Parallel and Distributed Systems,2012,23(11):1045-9219 [24] Madduri K,Su J,Williams S,et al.Optimization of Parallel Particle-to-Grid Interpolation on Leading Multicore Platforms[J].IEEE Transactions on Parallel and Distributed Systems,2012,23(10):1045-9219 [25] Gonz'alez-V'elez H,Leyton M.A survey of algorithmic skeleton frameworks high-level structured parallel programming enablers[J].Softw.Pract.Exper,2010,40:1135-1160 [26] Góes L F W,Ioannou N,Xekalakis P,et al.Autotuning Skeleton-Driven Optimizations for Transactional Worklist Applications[J].IEEE Transactions on Parallel and Distributed Systems,2012,23(12):1045-9219 [27] Scalosub G,Marbach P,Liebeherr J.Buffer Management forAggregated Streaming Data with Packet Dependencies[J].IEEE Transactions on Parallel and Distributed Systems,2013,24(3):1045-9219 [28] Nugteren C,Custers P,Corporaal H.Algorithmic Species:AClassification of Affine Loop Nests for Parallel Programming[J].ACM Transactions on Architecture and Code Optimization,2013,40(4) [29] S′aez S,Salvadory I.Exploiting Parallelism in Multi-View Systems using UML Activity Diagrams and OpenMP[C]∥2010 Workshops on Database and Expert Systems Applications.2010 [30] 王科特,王力生.信号实时采集的最佳并行线程数的研究[J].计算机应用,2011,31(10):2593-2596 [31] Peng Li-zhi,Yang Bo.A parallel evolving algorithm for flexible neural tree[J].Parallel Computing,2011,37:653-666 [32] Ovatmana T,Weigertb T.Exploring implicit parallelism in class diagrams[J].The Journal of Systems and Software, 2011,84:821-834 [33] Yang Chao-tung,Huang Chih-lin.Hybrid CUDA,OpenMP,and MPI parallel programming on multicore GPU clusters[J].Computer Physics Communications,2011,182:266-269 [34] Redondo J L,García I,Ortigosa P M.Parallel evolutionary algorithms based on shared memory programming approaches[J].J Supercomput,2011,58:270-279 [35] Sui Yang-yi,Lin Jun,Zhang Xiao-tuo.An Automated Refacto-ring Tool for Dataflow Visual Programming Language[J].ACM SIGPLAN Notices,2008,43(4) [36] Whitleya K N,Novick L R.Evidence in favor of visual representation for the dataflow paradigm:An experiment testing LabVIEW’s comprehensibility[J].Int.J.Human-Computer Studies,2006,64:281-303 [37] Tekinerdogan B,Aksit M.Impact of Evolution of Concerns in the Model-Driven Architecture Design Approach[J].Electronic Notes in Theoretical Computer Science,2007,163:45-64 [38] Jazayeri M,Oberleitner J.Predicting Incompatibility of Trans-formations in Model-driven Development[J].Electronic Notes in Theoretical Computer Science,2005,127:129-137 [39] Daniluk A.Visual modeling for scientific software architecturedesign.A practical approach[J].Computer Physics Communications,2012,183:213-230 [40] Hloupisa G,Stavrakas I.WSN Open Source Development Platform:Application to Green Learning[J].Procedia Engineering,2011,25:1049-1052 [41] Jordan H F,Alaghband G.Fundamentals of Parallel Processing[M].Chi Li-hua,Liu Jie,ed.Beijing:Tsinghua University Press,Oct.2004 [42] Herlihy M,Shavit N.The Art of Mutiprocessor Programming[M].Jing Hai,Hu Kan,ed.Beijing :China Machine Press,Aug.2009 [43] Culler D E,Singh J P,Gupta A.Parallel Computer Architecture:A Hardware/Software Approach(Second Edition)[M].Li Xiao ming,et al.ed.Beijing:China Machine Press,Oct.2002 [44] Giacaman N,Sinnen O.Parallel Task for Parallelising Object-Oriented Desktop Applications[J].Int J Parallel Prog,2013(41):621-681 [45] Vianna E,Comarela G.Analytical Performance Models for MapReduce Workloads[J].Int J Parallel Prog,2013(41):495-525 |
No related articles found! |
|