几种改进分布式共享存储器系统性能的算法  

Several Algorithms for Improving System Performance in Distributed Shared Memory Systems

作　　者：陈智勇[1] 

机构地区：[1]桂林电子工业学院计算机系,广西桂林541004

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

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.

关 键 词：卷回恢复 算法 分布式共享存储器 并行计算机 

分 类 号：TP333[自动化与计算机技术—计算机系统结构] TP301.2[自动化与计算机技术—计算机科学与技术]