论文信息 - An Implementation and Evaluation of a Distributed Shared-Memory System on Workstation Clusters Using Fast Serial Links

An Implementation and Evaluation of a Distributed Shared-Memory System on Workstation Clusters Using Fast Serial Links

We summarize an implementation of a distributed sharedmemory system on a workstation cluster. In this paper, we introduce fast serial links called Serial Transparent Asynchronous First-in Firstout Link (STAFF-Link). By using these links we construct a parallel processing system based on the workstation cluster. In the workstation cluster, a distributed shared-memory mechanism is utilized for interprocess communication with software controlled cache. We evaluate the performance of the system for several applications.

[1] Anoop Gupta,et al. SPLASH: Stanford parallel applications for shared-memory , 1992, CARN.

[2] Charles L. Seitz,et al. Myrinet: A Gigabit-per-Second Local Area Network , 1995, IEEE Micro.

[3] Paul Hudak,et al. Memory coherence in shared virtual memory systems , 1989, TOCS.

[4] Ruby B. Lee,et al. Tempest: a substrate for portable parallel programs , 1995 .

[5] Takashi Matsumoto,et al. An I/O network architecture of the distributed shared-memory massively parallel computer JUMP-1 , 1997, ICS '97.

[6] Vaidy S. Sunderam,et al. PVM: A Framework for Parallel Distributed Computing , 1990, Concurr. Pract. Exp..

[7] Daniel E. Lenoski,et al. Scalable Shared-Memory Multiprocessing , 1995 .

[8] Message P Forum,et al. MPI: A Message-Passing Interface Standard , 1994 .

[9] Alan L. Cox,et al. TreadMarks: Distributed Shared Memory on Standard Workstations and Operating Systems , 1994, USENIX Winter.