摘要
检查点技术,也称为“回溯恢复”,是软件容错的重要手段,它主要用于保存和恢复程序的运行状态。在分布式计算和并行计算系统中有十分重要的作用。该文从减少检查点的开销角度,对分布式系统检查点算法中关于程序卷回时文件系统状态的恢复问题进行了分析讨论和进一步的研究。
Checkpointing technology,which is called backout recovery,is used for conserving and reconvering the running status of program,and it is the important method for fault-tolerance of software.It also plays significant role in the distributed and parallel computing system.From the perspective of reducing the overhead of checkpointing,the paper analyses and discusses the reconversion problem of the status of file system about program recoiling of checkpointing algorithms in the distributed systems.
出处
《计算机工程与应用》
CSCD
北大核心
2002年第17期131-134,共4页
Computer Engineering and Applications
关键词
分布式系统
检查点算法
程序卷回
文件系统
状态恢复
计算机网络
Checkpointing,distributed systems ,consistent checkpointing algorithms ,file system status reconversion
