Abstract | ||
---|---|---|
Non-blocking multithreaded architectures have been proposed as an effective means to overlap computation and communication in distributed memory systems. However, in these models, communication latency could only be hidden from computation as long as there are enough ready threads. We describe our I-Structure Software Cache (ISSC) run-time system, which takes advantage of the global data locality in these models without adding any specific hardware support. Our ISSC provides a communication latency reduction technique in non-blocking multithreaded architectures while maintaining the ability of tolerating communication latency. Our simulation results show that our ISSC run-time system dramatically decreases network traffic by caching remote requests and also achieves 70% to 95% improvement on the system performance (speedup) |
Year | DOI | Venue |
---|---|---|
1997 | 10.1109/ISPAN.1997.645058 | ISPAN |
Keywords | Field | DocType |
exploiting global data locality,i-structure software cache,network traffic,cache storage,distributed memory systems,remote request caching,parallel architectures,virtual machines,system performance,global data locality,communication latency reduction technique,nonblocking multithreaded architectures,multithreaded architectures,multithreaded architecture,simulation,issc runtime system,performance evaluation,run-time system,effective mean,communication latency,runtime system,memory system,tolerating communication latency,computation,computer architecture,computational modeling,distributed computing,distributed memory,parallel processing,hardware,switches | Locality,Virtual machine,Software cache,Latency (engineering),Computer science,Parallel computing,Thread (computing),Distributed memory systems,Computation,Speedup | Conference |
ISSN | ISBN | Citations |
1087-4089 | 0-8186-8259-6 | 3 |
PageRank | References | Authors |
0.42 | 7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wen-Yen Lin | 1 | 58 | 11.47 |
Jean-Luc Gaudiot | 2 | 938 | 121.51 |