期刊文献+

公平消息传递并行程序设计的Petri网方法 被引量:1

Fair Message-passing Parallel Program Design with Petri Net
下载PDF
导出
摘要 公平性是并行程序正确的主要标准之一,但是并行程序执行的不确定性导致公平性验证较为困难。针对消息传递模式的并行程序,在建立相应Petri网模型的基础上,首先证明了并行程序的公平性与其Petri网模型的公平性之间的关系,然后提出了利用T-不变量判断并行程序公平性,以及通过添加控制库所实现并行程序公平的算法,最后指出了将改造后的模型转换为公平并行程序的方法,并用一个实例进行了验证。 Fairness is one of the main correct criterions of parallel programs, but the uncertainty of its execution leads to difficult verification. On the basis of modeling the message-passing parallel program with Petri nets, it was proved that the Petri net model of a fair parallel program is fair and has a unique support of T-invariant. In succession, the algorithm to discover the starvation, and ensure the fairness by appending control places was given. Finally, the transforming method from the reconstructed Petri net to a fair parallel program was discussed, and an example composed of two consumers and one producer was used to demonstrate the power of the method. This method can be used during the algorithm design period to improve programming efficiency.
出处 《系统仿真学报》 CAS CSCD 北大核心 2009年第13期3933-3936,3940,共5页 Journal of System Simulation
基金 国家自然科学基金(60673053 60773034) 山东省泰山学者建设工程基金项目 山东省中青年科学家科研奖励基金博士基金项目(2006BS01019) 山东科技大学春蕾计划项目(2008BZC018)
关键词 消息传递并行程序 公平性 PETRI网 T-不变量 控制库所 message-passing parallel program fairness Petri net T-invariant control place
  • 相关文献

参考文献4

二级参考文献18

  • 1陈曦,高仲仪.静态检测Ada任务死锁的Petri网方法[J].计算机研究与发展,1997,34(S1):145-149. 被引量:1
  • 2ChenWang,YinLiu,ChangjunJiang,ZhaoqingZhang.A Performance Analysis Tool for PVM Parallel Programs[J].计算机工程与应用,2004,40(29):103-105. 被引量:1
  • 3李代平,罗寿文,张信一,方海翔.一个网络并行计算新平台[J].计算机工程与设计,2005,26(1):24-26. 被引量:3
  • 4崔焕庆,吴哲辉.MPI程序的Petri网模型及其动态性质(英文)[J].系统仿真学报,2006,18(9):2455-2460. 被引量:5
  • 5布奇 G 麦中凡 梁南元译.Ada软件工程[M].北京:科学普及出版社,1986..
  • 6TotalView[EB/OL].http://www.etnus.com/TotalView,2006.
  • 7Kacsuk P.Performance visualization in the GRADE parallel programming environment[C].Proceedings of the 4th International Conference/Exhibition on High Performance Computing in Asia-Pacific Region.Los Alamitos,USA:IEEE Computer Society,2000:446-450.
  • 8Shende S,Malony A D.The TAU parallel performance system[J].International Journal of High Performance Computing Applications,2006,20(2):287-331.
  • 9Ferscha A,Johnson J.N-MAP:An environment for the performance oriented development process of efficient distributed programs[J].Future Generation Computer Systems,2000,16:571-584.
  • 10Bourgeois J,Spies F Chronos.A performance characterization tool inside the EDPEPPS toolset[J].The Journal of Supercomputing,2000,15(2):123-140.

共引文献8

同被引文献3

引证文献1

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