期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

JAPS中的任务划分与粒度控制 被引量：2

Task Partitioning and Granularity Controlling Orithmetic Used in JAPS
下载PDF
导出
摘要 程序并行化为了达到缩短程序运行时间的目的 ,需要关注两方面的问题 :1,任务之间并行性的挖掘 ;2 ,粒度控制 ;如何协调这两方面的关系以及如何自动的完成这个任务是现在工作的重点 .本文详细讨论了如何在 CDDG图上决定合适的任务粒度 ,提出基于关键路径的任务粒度控制理论 ;并描述了在 JAPS中实现的自动任务划分与粒度控制算法 .涉及的主要问题包括结点的分布 ,任务的粒度控制 ,任务的划分 。 Parallelizing Compiler requires solve two problems to shorten the programs running time. How to gain the best parallelism between tasks; * Grain size problem; Now, how to balance this two problems and how to finish this work automatic is most studied. This paper discusses how to decide the appropriate task size in CDDG graph and presents the theoretics of task size controlling basing on the key path. Then, we describe the automatic task partitioning and granularity controlling arithmetic used in JAPS. The including problems are the distributing of node, the task size controlling and the partition of task. Last we analyse the result in JAPS.
作者 龚心宇 于勐 李晓明 陈道蓄 谢立
机构地区 南京大学计算机软件新技术国家重点实验室 南京大学计算机科学与技术系
出处 《小型微型计算机系统》 CSCD 北大核心 2002年第2期180-184,共5页 Journal of Chinese Computer Systems
基金 国家 8 63基金资助
关键词 CDDG图 任务划分 粒度控制 程度并行化 JAPS CDDG graph parallelism task partition task size controlling
分类号 TP31 [自动化与计算机技术—计算机软件与理论]
  • 相关文献

参考文献2

共引文献4

同被引文献13

  • 1Liu C L, Layland J. Scheduling algorithms for multi programming in a hard-real-time environment [J]. Journal of the Association for Computing Machinery, 1973, 20(1): 46-61.
  • 2Zhou Benhai, Qing Jianzhong, Lin Shukuan. Re- search on synthesis parameter real-time scheduling al- gorithm on multi-core architecture[C] // Control and Decision Conference. Gulin: IEEE, 2009: 5116- 5120.
  • 3Bastoni A, Brandenburg B B, Anderson J H. An em- pirical comparison of global, partitioned, and clus- tered multiprocessor EDF schedulers[C]//Real-Time Systems Symposium (RTSS), 2010 IEEE 31st. San Diego.. IEEE, 2010:14-24.
  • 4Liu Jiahai, Yang Maolin. Task scheduling of real- time systems on multi core embedded processor[C]// International Conference on Intelligent Systems and Knowledge Engineering (ISKE). Hangzhou: IEEE, 2010: 580-583.
  • 5Davis R, Burns A. Improved priority assignment for global fixed priority preemptive scheduling in multi- processor real-time systems[J]. Real-Time Systems, 2011, 47(1): 1-40.
  • 6Bozdag D, Ozguner F, Catalyurek U V. Compaction of schedules and a two-stage approach for duplication- based DAG scheduling[J]. IEEE Transactions on Parallel and Distributed Systems, 2009, 20(6):857- 871.
  • 7Lin Siming, Chen Xueqi. BC-GA: a graph partitio ning algorithm for parallel simulation of internet ap plications[C]//16th Euromicro Conference on Paral lel, Distributed and Network-Based. Toulouse IEEE, 2008: 358-365.
  • 8Zhou Lan, Sun Shixin. A genetic scheduling algorithm based on knowledge for multiprocessor sys- tem, in communications, circuits and systems[C]// ICCCAS 2007. Fukuoka: IEEE, 2007: 900-904.
  • 9Cong L, Anderson J H. Scheduling suspendable, pipelined tasks with non-preemptive sections in soft real-time multiprocessor systems[C]//16th IEEE Real-Time and Embedded Technology and Applica- tions Symposium (RTAS). Stockholm: IEEE, 2010: 23-32.
  • 10Andersson B, Pinho L M. Implementing multicore real-time scheduling algorithms based on task split- ting using Ada 2012 [C]///Reliable Software Techn- ologiey-Ada-Europe 2010. Berlin/Heidelberg : Springer, 2010: 54-67.

引证文献2

二级引证文献1

小型微型计算机系统
2002年 第2期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部