Multiprocessor architectures and physical law

P. Vitányi
{"title":"Multiprocessor architectures and physical law","authors":"P. Vitányi","doi":"10.1109/PHYCMP.1994.363703","DOIUrl":null,"url":null,"abstract":"We show that all highly symmetrical interconnection topologies for multiprocessors with low diameter require very long interconnect lengths. Therefore, such multicomputers do not scale well in the physical world with 3 dimensions. On the other hand, highly irregular (random) interconnection topologies have a very large subgraph of diameter two and therefore also require very long interconnect lengths. Hence the only scaling topologies for future massively parallel computers are high diameter regular ones, like mesh networks. The techniques used are symmetry properties in terms of orbits of automorphism groups of graphs, and a modern notion of randomness of individual objects, Kolmogorov complexity.<<ETX>>","PeriodicalId":378733,"journal":{"name":"Proceedings Workshop on Physics and Computation. PhysComp '94","volume":"72 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1994-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings Workshop on Physics and Computation. PhysComp '94","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PHYCMP.1994.363703","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 9

Abstract

We show that all highly symmetrical interconnection topologies for multiprocessors with low diameter require very long interconnect lengths. Therefore, such multicomputers do not scale well in the physical world with 3 dimensions. On the other hand, highly irregular (random) interconnection topologies have a very large subgraph of diameter two and therefore also require very long interconnect lengths. Hence the only scaling topologies for future massively parallel computers are high diameter regular ones, like mesh networks. The techniques used are symmetry properties in terms of orbits of automorphism groups of graphs, and a modern notion of randomness of individual objects, Kolmogorov complexity.<>
多处理器体系结构和物理定律
我们证明了所有高度对称的低直径多处理器互连拓扑都需要非常长的互连长度。因此,这种多计算机在三维物理世界中不能很好地扩展。另一方面,高度不规则(随机)互连拓扑具有非常大的直径为2的子图,因此也需要非常长的互连长度。因此,未来大规模并行计算机的唯一可伸缩拓扑是大直径的规则拓扑,如网状网络。使用的技术是图的自同构群的轨道的对称性,以及单个对象的随机性的现代概念,Kolmogorov复杂度
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信