Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN)最新文献_第4页

Massively Parallel Processing Project as a Priority Area of Research for the Ministry of Education 大规模并行处理项目作为教育部重点研究领域

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367141

Hidehiko Tanaka

引用次数: 0

A logic semantics for a class of nondeterministic concurrent constraint logic programs 一类不确定并发约束逻辑程序的逻辑语义

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367181

Ho-fung Leung, Bo-Ming Tong

引用次数: 0

Parallel relational database algorithms revisited for range declustered data sets 重新审视了范围聚类数据集的并行关系数据库算法

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367168

E. Schikuta

引用次数: 2

Transit: reliable high speed interconnection technology 传输:可靠的高速互连技术

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367180

T. Knight

引用次数: 1

Cenju-3 parallel computer and its application to CFD Cenju-3并行计算机及其在CFD中的应用

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367184

K. Muramatsu, S. Doi, T. Washio, T. Nakata

引用次数: 7

A parallel algorithm for the single step searching problem 单步搜索问题的并行算法

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367138

J. S. Lin, F. Hsu, Richard C. T. Lee

引用次数: 0

Recursive circulant: a new topology for multicomputer networks (extended abstract) 递归循环:多计算机网络的一种新拓扑(扩展摘要)

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367162

Jung-Heum Park, Kyung-Yong Chwa

引用次数: 98

A shortest path algorithm for banded matrices by a mesh connection without processor penalty 一种无处理器损失的网格连接带状矩阵最短路径算法

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367151

Aohan Mei, Y. Igarashi

引用次数: 1

Processing nested loop structure with data-flow dependence on a CAM-based processor HAPP 处理数据流依赖于基于cam的处理器HAPP的嵌套循环结构

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367156

K. Lu, K. Tamaru

引用次数: 0

Building a better butterfly: the multiplexed metabutterfly 建造一个更好的蝴蝶:多路元蝴蝶

Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN) Pub Date : 1994-12-14 DOI: 10.1109/ISPAN.1994.367163

F. Chong, E. Brewer, F. Leighton, T. Knight

{"title":"Building a better butterfly: the multiplexed metabutterfly","authors":"F. Chong, E. Brewer, F. Leighton, T. Knight","doi":"10.1109/ISPAN.1994.367163","DOIUrl":"https://doi.org/10.1109/ISPAN.1994.367163","url":null,"abstract":"Multistage networks are important in a wide variety of applications. Expander-based networks, such as multibutterflies, are a tremendous improvement over traditional butterflies in both fault and congestion tolerance. However, multibutterflies cost at least twice as much in chips and wiring as butterflies. It is also impossible to build large multibutterflies due to their wiring complexity. We show that we can build an expander-based network that has comparable cost to a butterfly with the same number of endpoints, yet has substantially better fault and congestion performance. Specifically, we introduce a hierarchical construction that dramatically reduces wiring complexity and makes large expanders buildable. We are able to exploit the hierarchical structure to find large numbers of logical wires to multiplex over a smaller number of physical wires. Since many of the wires in an expander-based network are used to provide alternate paths, not useful bandwidth, substantial multiplexing can be done without significantly degrading performance. We present simulation results to support our conclusions. In comparing a butterfly with the comparable 2-to-1 multiplexed metabutterfly, we found that the metabutterfly performed better by nearly a factor of two on random traffic and greater than a factor five on worst-case traffic.<<ETX>>","PeriodicalId":142405,"journal":{"name":"Proceedings of the International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115682449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8