期刊文献+

基于排序机制的全序广播算法分类与性能分析 被引量:1

Classification and performance of total order broadcast and multicast algorithms based on the mechanism of ordering
下载PDF
导出
摘要 已有的全序广播和多播算法多达近五十种,为了便于人们根据需要从众多算法中进行选择,需要对这些算法进行明确分类。但现有的分类方法都存在着完备性较差,且难以理解等问题。基于排序机制,从全序广播和多播算法的通信记录、优先权、动态序列器、静态序列器和目的方一致五个方面,给出了一种针对全序广播与多播算法的新系统分类方法。同时,还从定性、定量两个方面对分类的各类算法进行了规范、应用模型环境以及性能的分析与评测。使用新的分类方法以及测评结果,可以在现实应用中很方便地从现有算法集中根据需要选择合适的算法。 So far, there are almost are almost fifty algorithms on total order broadcast and multicast, which makes it too difficult to select a best suited algorithm to a given application context. Though some attempts have been made at classifying and comparing these algorithms recently, none is comprehensive, and hence lacks generality. A novel classification system is proposed by using the ordering mechanisms. In this system, five classes of total order broadcast and multicast algorithms are defined, I.e. communication history based algorithm, privilege-based algorithms, dynamic sequencer algorithm, static sequencer algorithm and destination agreement algorithm. Then existing algorithms are related to these classes. Also, the performances of each class are analyzed qualitatively and quantitatively. With this classification method and the analytical result, it is very convenient to choose an appropriate algorithm for a specific a specific application.
出处 《系统工程与电子技术》 EI CSCD 北大核心 2005年第1期149-154,共6页 Systems Engineering and Electronics
基金 国家自然科学基金重大研究计划项目(90204012) 国家高技术研究发展计划("836"计划)(2002AA143021) 教育部优秀青年教师资助计划 教育部科学技术重点研究项目 总装备部武器装备基金项目资助课题。
关键词 全序广播算法和全序多播算法 排序机制 分类 性能分析 total order broadcast and multicast algorithms ordering mechanism classification performance analysis
  • 相关文献

参考文献53

  • 1Lamport L. Time, clocks, and the ordering of events in a distributed system[J]. Communications of the ACM, 1978, 21(7) :558 - 565.
  • 2Schneider F B. Implementing fault - tolerant services using the state machine approach: a tutorial[J]. ACM Computing Surveys, 1990:22(4):299 - 319.
  • 3Chockler G V, Huleihel N, Dolev D. An adaptive total ordered multicast protocol that tolerates partitions[C]. Puerto Vallarta, Mexico, In Proceedings of the 17th ACM Symposium on Principles of Distributed Computing (PODC- 17), 1998:237 -246.
  • 4Birman K P and Joseph T A. Reliable communication in the presence of failures[J]. ACM Transactions on Computer Systems, 1987, 5(1) :47 -76.
  • 5Luan S W, Gligor V D. A fault - tolerant protocol for atomic broadcast [J]. IEEE Transactions on Parallel and Distributed Systems, 1990, 1(3) : 271-285.
  • 6Anceaume E. A comparison of fault - tolerant atomic broadcast protocols [C]. Lisbon, Portugal, In Proceedings of the 4th IEEE Computer Society Workshop on Future Trends in Distributed Computing Systems (FTDCS-4), 1993: 166-172.
  • 7Cristian F, de Beijer R, Mishra S A. performance comparison of asynchronous atomicbroadcast protocols[J]. Distributed Systems Engineering,1994, 1(4):177-201.
  • 8Peterson L L, Buchholz N C, Schlichting R D. Preserving and using context information in interprocess communication[J]. ACM Transactions on Computer Systems, 1989, 7(3) :217 - 246.
  • 9Moser L E, Melliar- Smith P M, Agrawala. V Total ordering algorithms. In ACM Annual Computer Science Conference[C]. Preparing for the 21st Century, 1991:375 - 380.
  • 10Moser L E, Melliar-Smith P M, Agrawala V. Asynchronous fault- tolerant total ordering algorithms[J]. SIAM Journal on Computing, 1993,22(4) :727 - 750.

同被引文献7

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