期刊文献+

分布式仿真系统socket迁移 被引量:3

Socket Migration in Distributed Simulation System
下载PDF
导出
摘要 在分布式仿真系统进程迁移中,除了要保存、恢复仿真进程本身的状态,进程之间通信状态的保存与恢复也是关键。通信状态主要包括两部分:进程IP地址以及进程间的交互数据。我们通过截获系统调用的方式,在用户空间完成了上述通信状态保存与恢复,实现了socket迁移。其中连接错误的监控通过心跳机制实现;进程的IP地址及交互消息被修改的系统调用透明的保存在checkpoint文件中;基于保存的上述信息,我们实现了联邦成员和RTI之间通信关系的恢复。测试结果表明该系统可以实现socket通信关系的迁移。 In the process migration of distributed simulation system, except saving and restoring the state of processes, it is also critical to save and restore the state of network communication between the simulation processes. The network communication states mainly consist of two parts, that is, the IP address and interactive data. Socket migration in user space was realized by intercepting and revising corresponding system calls. In the method, heartbeat mechanism monitors the connection error. Further, the system calls was revised, which could store the IP addresses and interactive data in checkpoint files. Based on the saved information, the recovery of communication relationship between federates and RTI was realized. The test result shows that the system can successfully achieve the migration of the socket communication relationship.
出处 《系统仿真学报》 EI CAS CSCD 北大核心 2006年第2期370-372,共3页 Journal of System Simulation
基金 国防预研基金资助(51404010403KG0155)
关键词 分布式仿真 RTI 容错 socket迁移 distributed simulation RTI fault tolerance socket migration
  • 相关文献

参考文献8

  • 1刘云生,张童,张传富,查亚兵.基于网格的分布式仿真系统容错机制[J].国防科技大学学报,2005,27(1):35-38. 被引量:3
  • 2黄柯棣等编著..系统仿真技术[M].长沙:国防科技大学出版社,1998:362.
  • 3Mootaz Elnozahy,Lorenzo Alvisi,Yi-Min Wang,David B.Johnson.A Survey of Rollback-Recovery Protocols in Message-Passing Systems [R].Technical report CMU-CS-99-148,School of Computer Science,Carnegie Mellon University,1999. 被引量:1
  • 4Dejan Milojicic,Fred Douglis,Songnian Zhou,etc.Process Migration [R].HP Labs,AT&T Labs-Research,TOG Research Institute,EMC,and University of Toronto and Platform Computing.1999. 被引量:1
  • 5蒋江..异构集群系统中基于进程迁移机制的负载平衡算法的研究[D].国防科学技术大学,2002:
  • 6Victor C Zandy,etc.Reliable Network Connections [DB/OL].Computer Sciences Department,University of Wisconsin-Madison. 被引量:1
  • 7JohnGoerzen.Linux Programming Bible [M].北京:电子工业出版社,2000.. 被引量:1
  • 8WRichardStevens.UNIX环境高级编程[M].北京:机械工业出版社,2003.. 被引量:1

二级参考文献9

  • 1金士尧 马民.HLA分布式仿真中容错机制研究[A]..第十届全国容错计算学术会议[C].,2003.. 被引量:2
  • 2GoerzenJ.Linux编程宝典[M].北京:电子工业出版社,2000.. 被引量:1
  • 3Elnozahy M, Alvisi L, Wang Y M, et al. A Survey of Rollback-recovery Protocols in Message-passing Systems[R]. Technical Report CMU-CS-99-148, School of Computer Science, Carnegie Mellon University, June 1999. 被引量:1
  • 4Milojicic D, Douglis F, Zhou S,et al. Process Migration[R]. HP Labs, AT&T Labs-Research, TOG Research Institute, EMC, and University of Toronto and Platform Computing,Feb,1999. 被引量:1
  • 5Stelling P, Foster I, Kesselman C. A Fault Detection Service for Wide Area Distributed Computations[J]. 0-8186-8579-4/98, IEEE,1998. 被引量:1
  • 6Dahmann J S. The High Level Architecture and Beyond:Technology Challenges[A]. In Proceedings of 13^th Workshop on Parallel and Distributed Simulation[C], 1999. 被引量:1
  • 7Kiesling T. Fault-tolerant Distributed Simulation: A Position Paper[R].2003. 被引量:1
  • 8Lüthi J,Berchtold C. Concepts for Dependable Distributed Discrete Event Simulation[A]. In Proceedings of the International European Simulation Multi-conference[C], 2000. 被引量:1
  • 9刘鹏.网格计算[M].北京:清华大学出版社,2003.. 被引量:1

共引文献2

同被引文献28

引证文献3

二级引证文献15

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