Local-area multiprocessor: surpassing clusters

D. Gustavson, Qiang Li
{"title":"Local-area multiprocessor: surpassing clusters","authors":"D. Gustavson, Qiang Li","doi":"10.1109/CMPCON.1995.512406","DOIUrl":null,"url":null,"abstract":"Clusters, or networks, of workstations (NOW) have proven their value for certain classes of applications, those that can be efficiently parallelized with coarse granularity requiring relatively little intertask communication. However clusters of workstations have been ineffective for those classes of applications that require fine-grained parallelization or too much intertask communication, because of the high overhead and poor performance of present communication mechanisms. A new international standard reduces this cost by several orders of magnitude compared to present networks or proposed future networks/channels like FibreChannel or ATM. This new technology implements a \"Local-Area Multiprocessor\" with very high performance message passing and distributed shared memory.","PeriodicalId":415918,"journal":{"name":"Digest of Papers. COMPCON'95. Technologies for the Information Superhighway","volume":"21 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1995-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Digest of Papers. COMPCON'95. Technologies for the Information Superhighway","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CMPCON.1995.512406","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Clusters, or networks, of workstations (NOW) have proven their value for certain classes of applications, those that can be efficiently parallelized with coarse granularity requiring relatively little intertask communication. However clusters of workstations have been ineffective for those classes of applications that require fine-grained parallelization or too much intertask communication, because of the high overhead and poor performance of present communication mechanisms. A new international standard reduces this cost by several orders of magnitude compared to present networks or proposed future networks/channels like FibreChannel or ATM. This new technology implements a "Local-Area Multiprocessor" with very high performance message passing and distributed shared memory.
局部多处理器:超越集群
工作站的集群或网络(NOW)已经证明了它们对某些应用程序的价值,这些应用程序可以用粗粒度有效地并行化,需要相对较少的任务间通信。然而,工作站集群对于那些需要细粒度并行化或过多任务间通信的应用程序来说是无效的,因为目前的通信机制开销高,性能差。与现有网络或拟议的未来网络/通道(如FibreChannel或ATM)相比,新的国际标准将这一成本降低了几个数量级。这种新技术实现了具有高性能消息传递和分布式共享内存的“局域多处理器”。
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