Collaborative Research: NeTS-FIND: Postcards from the Edge: A Cache-and-Forward Architecture for the Future Internet

提案编号：0626959 PI：Roy Yates研究所：罗格斯大学提案编号：0626874 PI：James Kurose机构：马萨诸塞大学标题：Collaborative NetTS-FIND：来自边缘的明信片：一种面向未来Internet的高速缓存转发体系结构一种前向网络体系结构，利用存储设备不断降低的成本和不断增加的容量，为可能通过有线连接的终端主机提供统一、高效的传输服务或无线的;静态的、移动的和/或间歇断开的;以及资源丰富或贫乏。 此架构的基础是传输层服务，它以逐跳存储和转发的方式处理大型文件。 为了实现这种体系结构，本项目设计、实现和评估了一种新的网络体系结构，该体系结构包括以下要素：（1）可靠的逐跳传输大文件;（2）推拉体系结构，用于向有线网络和从有线网络机会性地传送文件;（3）增强命名，为移动的终端提供位置信息;以及（4）流行内容的分布式高速缓存，以使对等文件共享成为一流的服务，并实现高效可靠的多播。该项目的实验评估阶段，广泛使用的PlanetLab的基础上，ORBIT无线电网格测试平台和真实的世界环境支持的未来实验系统，检查网络性能权衡各种潜在的协议设计选择和使用/流量的情况。该项目的技术优势在于建筑设计的互补组合，量化建筑组件性能优势的宏观模型的开发，以及关键建筑创新的原型实施和实验验证。 这个项目的更广泛的影响是它对未来互联网的一个或多个协议体系结构的定义和选择的贡献，导致新的服务和应用的价值，科学和商业最终用户。

Proposal Number: 0626959PI: Roy Yates Institution: Rutgers UniversityProposal Number: 0626874PI: James Kurose Institution: University of MassachusettsTitle: Collaborative NeTS-FIND: Postcards from the Edge: A Cache-and-Forward Architecture for the Future Internet AbstractThis project develops a cache-and-forward network architecture that exploits the decreasing cost and increasing capacity of storage devices to provide unified and efficient transport services to end hosts that may be wired or wireless; static, mobile, and/or intermittently disconnected; and either resource rich or poor. Fundamental to this architecture is a transport layer service that operates in a hop-by-hop store-and-forward manner with large files. To realize this architecture, this project designs, implements and evaluates a new network architecture that incorporates the following elements: (1) reliable hop-by-hop transport of large files; (2) push-pull architecture for opportunistic delivery of files both to and from the wired network; (3) enhanced naming to provide location information for mobile terminals; and (4) distributed caching of popular content to make peer-to-peer file sharing a first-class service and to enable efficient reliable multicast. The experimental evaluation phase of the project, based on extensive use of PlanetLab, the ORBIT radio grid testbed and real world environments supported by future experimental systems, examines network performance trade-offs for various potential protocol design choices and usage/traffic scenarios. The technical merit of this project is in the complementary mix of architectural design, the development of macroscopic models quantifying the performance benefits of architectural components, and a prototype implementation and experimental validation of key architectural innovations. The broader impact of this project is its contribution towards the definition and selection of one or more protocol architectures for the future Internet, leading to new services and applications of value to both scientific and commercial end users.