摘要
一、引言
分布式共享内存(Distributed Shared Memory,DSM)是并行处理中的一种关键技术.它为程序员提供了一个逻辑上统一的虚拟地址空间,任何一个处理机都可以对这一地址空间直接进行读写访问.其中一致性模型、一致性协议、粒度是影响DSM性能的重要因素.
DSM中物理内存分布在多个节点上,数据一般采用复制的方式来实现共享内存的抽象,因此如何在保证内存一致性前提下尽量提高效率成为DSM的一个重要研究内容.一致性模型[5]本质上是内存系统与并行程序之间关于内存访问的协议.严格一致性模型包括原子一致性、顺序一致性等.严格的一致性模型不利于开发程序的并行性,因此出现了很多松懈一致性模型,包括弱一致性、释放一致性、域一致性(Scope Consistency)和单项一致性(Entry Consistency)等.
We present a hierarchical DSM scheme named GDSM. In our solution,sub-tasks are to be mapped to groups,which apply different techniques based on the distinct characteristics of inter-group and intra-group data sharing. A serial of alternatives are used with respect to the intra-group and inter-group property of this mechanism: (1 )Consistency model:release consistency vs. scope consistency; (2) Coherence protocol: multiple-writer and write-update protocol vs. single-writer and write-invalidate protocol;(3)Granularity:fine-grain vs. coarse-grain. Our novel strategy to combine grouping with consistency,data conversion and granularity switch cuts down the overhead of consistency maintenance,increases the parallelism between groups and thus promotes system performance. We also put forward the architecture of Protocol Engine implementing the GDSM scheme,the engine that achieves message-forwarding transparency ,clear hierarchical structure and encapsulation of group.
出处
《计算机科学》
CSCD
北大核心
2001年第1期39-44,共6页
Computer Science
基金
国家自然科学基金(NSF#69803005)
关键词
DSM模型
层次结构
并行处理
软件
DSM .Group ,Consistency model ,Granularity ,Protocol Engine