A Scalable High-performance Router Platform Supporting Dynamic Service Extensibility On Network and Host Processors

L. Ruf, R. Keller, B. Plattner
{"title":"A Scalable High-performance Router Platform Supporting Dynamic Service Extensibility On Network and Host Processors","authors":"L. Ruf, R. Keller, B. Plattner","doi":"10.1109/ICPS.2004.6","DOIUrl":null,"url":null,"abstract":"Emerging network services such as transcoding and encryption need application-specific handling of data streams within the network, thus requiring enormous computational capabilities on routers to process packets at link speed. Recently appeared Network Processors (NPs) are able to significantly increase the available processing capacities on a router by a chip-multi-processor architecture. Embedded within the network interface card, NPs provide several code-extensible processors with different capabilities located at various layers. However, the increase in processing capacity comes at the cost of a higher complexity to program and control various processor hierarchies provided on the router. In this paper, we introduce the model of active network nodes built of a processor hierarchy together with a component-based service model. We present the architecture of PromethOS NP which is a modular framework that manages and controls a network node with a multitude of processors in a scalable way. Specifically, we describe the mechanisms required to dynamically configure multiple processors organized in a hierarchical fashion such that they provide a new network service on behalf of applications. As a proof of concept, we have implemented a service framework for PromethOS NP. Our implementation is based on a network interface card with an embedded IBM PowerNP 4GS3 and an Intel Xeon processor, offering programmability on a three-tier hierarchy of processors.","PeriodicalId":222266,"journal":{"name":"The IEEE/ACS International Conference on Pervasive Services","volume":"22 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"The IEEE/ACS International Conference on Pervasive Services","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPS.2004.6","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14

Abstract

Emerging network services such as transcoding and encryption need application-specific handling of data streams within the network, thus requiring enormous computational capabilities on routers to process packets at link speed. Recently appeared Network Processors (NPs) are able to significantly increase the available processing capacities on a router by a chip-multi-processor architecture. Embedded within the network interface card, NPs provide several code-extensible processors with different capabilities located at various layers. However, the increase in processing capacity comes at the cost of a higher complexity to program and control various processor hierarchies provided on the router. In this paper, we introduce the model of active network nodes built of a processor hierarchy together with a component-based service model. We present the architecture of PromethOS NP which is a modular framework that manages and controls a network node with a multitude of processors in a scalable way. Specifically, we describe the mechanisms required to dynamically configure multiple processors organized in a hierarchical fashion such that they provide a new network service on behalf of applications. As a proof of concept, we have implemented a service framework for PromethOS NP. Our implementation is based on a network interface card with an embedded IBM PowerNP 4GS3 and an Intel Xeon processor, offering programmability on a three-tier hierarchy of processors.
一个可扩展的高性能路由器平台,支持网络和主机处理器上的动态业务扩展
诸如转码和加密之类的新兴网络服务需要对网络中的数据流进行特定于应用程序的处理,因此需要路由器具有巨大的计算能力以链路速度处理数据包。最近出现的网络处理器(NPs)能够通过芯片多处理器架构显着增加路由器上的可用处理能力。NPs嵌入在网络接口卡中,提供了几个位于不同层的具有不同功能的代码可扩展处理器。然而,处理能力的增加是以更高的编程复杂性和控制路由器上提供的各种处理器层次为代价的。本文介绍了基于处理器层次结构的活动网络节点模型和基于组件的服务模型。我们介绍了PromethOS NP的体系结构,它是一个模块化框架,可以以可扩展的方式管理和控制具有众多处理器的网络节点。具体来说,我们描述了动态配置以分层方式组织的多个处理器所需的机制,以便它们代表应用程序提供新的网络服务。作为概念验证,我们已经为PromethOS NP实现了一个服务框架。我们的实现基于带有嵌入式IBM PowerNP 4GS3和Intel Xeon处理器的网络接口卡,在三层处理器层次结构上提供可编程性。
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:604180095
Book学术官方微信