NeTS-NR: Generalizing the Network Layer

该研究项目旨在开发一个能够在给定的拓扑上使用单一机制支持单播、组播和发布-订阅服务的综合网络层。这种服务将包括目前提议和使用的许多特殊目的方法(例如覆盖)，并将以更直接、更统一的方式支持新的应用和移动性。该方法基于使用分组中携带的谓词来识别分组目的地。换句话说，网络在终端系统上定义了一组谓词；每个分组都携带来自该集合的一个谓词，并进行了适当的编码，网络的任务是(在尽力的基础上)将分组传送到满足其目的地谓词的所有终端系统(以及尽可能少的其他系统)。主要的挑战是设计服务的注册、路由和转发算法，使系统能够扩展到支持比当前Internet大得多的网络。我们的目标是以与现有方法相当或更好的总体性能、管理和运营开销来实现这一目标。本研究项目将探索拓扑、转发效率和谓词语言之间的基本关系和权衡。它将开发在表达能力和效率之间取得适当平衡的谓词结构。最后，本项目将基于这些谓词结构设计、分析和评估高效的路由和转发算法和协议。研究结果有可能使新的网络能够支持更多种类的应用服务，同时避免阻碍当前系统的广域路由管理问题。

This research project aims to develop an integrated network layer capable of supporting unicast, multicast, and publish-subscribe services on a given topol-ogy using a single mechanism. Such a service would subsume many of the special-purpose approaches (e.g. overlays) currently being proposed and used, and would support novel applications and mobility in a more direct, unified way. The approach is based on the use of predicates carried in packets to iden-tify packet destinations. In other words, the network defines a set of predicates over end systems; each packet carries a predicate from this set, suitably en-coded, and the job of the network is to deliver the packet (on a best-effort basis) to all end systems that satisfy its destination predicate (and to as few others as possible). The major challenge is to design the enrollment, routing, and for-warding algorithms of the service in such a way that the system can scale up to support networks much larger than the current Internet. The goal is to achieve this with overall performance, administrative, and operational overhead compa-rable to or better than existing approaches. This research project will explore the fundamental relationships and tradeoffs among topology, forwarding effi-ciency, and predicate language. It will develop predicate structures that strike a suitable balance between expressive power and efficiency. Finally, the project will design, analyze and evaluate efficient routing and forwarding algorithms and protocols based on those predicate structures. Results of the research have the potential to enable new networks capable of supporting a greater variety of application services while avoiding problems with wide-area routing administra-tion that hamper the current system.