摘要
分布式共享存储器 (DSM)系统结合了共享存储器并行计算机编程容易和消息传递多计算机的可扩展性。在此介绍了 DSM系统中的检查点与卷回恢复算法、共享粒度的动态适应算法 ,以及各种算法的原理和算法对 DSM系统性能的影响。
Distributed Shared Memory(DSM) systems combine the ease of programming of shared memory parallel computers and the extension for the multi computers of message passing . This article presents several algorithms for checkpoint and rollback recovery, and the dynamic adaptation of sharing granularity in DSM systems. The principle of every algorithm, and the effect of every algorithm on DSM system performance are introduce.
出处
《桂林电子工业学院学报》
2001年第3期10-13,共4页
Journal of Guilin Institute of Electronic Technology
关键词
卷回恢复
算法
分布式共享存储器
并行计算机
distributed shared memory,checkpoint,rollback recovery,sharing granularity,algorithm
