ITR: Collaborative Research: Scalable Services for the Global Network

新应用程序的出现催生了许多向极简的互联网核心添加功能的尝试。然而，对Internet增强的采用要么很慢，要么完全失败，要么仅限于特殊用途的私有网络。这种失败的关键原因是可扩展性和可伸缩性：首先，IP网络没有设计成在互连层上可扩展。其次，新的网络层服务通常要求在核心网络基础设施中管理大量的状态信息，从而引入了可伸缩性瓶颈，加剧了日益增长的互联网现有的可伸缩性问题。今天，在Internet上开发和部署高级服务已经到了一个十字路口：添加新服务的努力很快就遇到了可伸缩性问题，而新服务的需求非常迫切，必须迅速和广泛地部署。研究目标是为互联网基础设施的三个基本组件（信息通信、复制和存储）中的每一个开发真正可扩展的服务。采用一种新的统一的方法来解决可伸缩性和复杂网络服务的看似冲突的需求，研究人员建议开发：可伸缩的性能可预测通信：通过可伸缩的基于边缘的架构为服务质量通信奠定了新的基础。2. 有效数据传播的可扩展多播：一个可扩展到许多自发形成的组的自组织多播基础设施。3. 下一代信息服务的可伸缩存储：一种使信息更接近用户并支持可伸缩第三方信息存储服务的基础设施。4. 可扩展服务的设计原则：在考虑经济模型、产业结构、理论和算法、工程和部署的情况下，在全球互联网中开发和部署可扩展服务的多方面方法。因此，该项目建议开发用于在全球Internet中部署可伸缩服务的体系结构和方法。该项目的影响将是为21世纪全球互联网的信息通信提供理论基础、基本架构和原型实现。这个项目的一个组成部分是成果的传播和标准组织的渗透，这些组织在这个项目中开发的概念，以及为未来互联网教育下一代工程师的创新方法。

The emergence of new applications has fostered a number of attempts to add functionalities to the minimalist Internet core. However, the adoption of enhancements to the Internet has either been slow, failed entirely, or limited to special-purpose private networks. The key reasons for this failure are extensibility and scalability: First, IP networks were not designed to be extensible at the internetworking level. Second, proposals for new network layer services often require that vast amounts of state information be managed in the core network infrastructure, thus, introducing scalability bottlenecks which exacerbate the existing scalability problem of the growing Internet. Today, the development and deployment of advanced services on the Internet has reached a crossroads: efforts to add new services have quickly encountered scalability problems, yet new services are in critical demand and must be rapidly and widely deployed. The research goal is to develop truly scalable services for each of the three fundamental components of the Internet's infrastructure: information communication, replication, and storage. Taking a new and unified approach to the seemingly conflicting requirements for scalability and sophisticated network services, the researchers propose to develop: 1. Scalable Performance-Predictable Communication: a new foundation for quality-of-service communication via a scalable edge-based architecture. 2. Scalable Multicast for Efficient Data Dissemination: a self-organizing multicast infrastructure scalable to many spontaneously-formed groups. 3. Scalable Storage for Next Generation Information Services: an infrastructure which brings informationcloser to users and enables scalable third-party information storage services. 4. Design Principles of Scalable Services: a multi-faceted approach for the development and deploymentof scalable services in the global Internet, under consideration of economic models, industrial structure,theories and algorithms, engineering, and deployment. Thus, this project proposes to develop architectures and methodologies for deploying scalable services in the global Internet. The impact of this project will be to provide the theoretical underpinnings, basic architecture, and a prototype implementation for the information communication of the global Internet of the 21st century. An integral part of this project is the dissemination of results and the infiltration of standard organizations with the concepts developed within this project, and innovative approaches to educate the next generation of engineers for the future Internet.