面向多自治域环境的信息中心网络缓存策略和机制研究

泛在缓存是众多信息中心网络(ICN)架构的基石。目前，关于泛在缓存的研究主要聚焦于通过缓存协同提高单个缓存网络的性能，却忽视了对网络的自治管理需求，而这对ICN的商业化和全球化至关重要。当把ICN拓展到多个自治域时，缓存不仅能降低网络提供商的流量，还能成为其商业关系谈判的筹码。因此，制定合理的缓存策略以实现与其他网络的商业契约关系，并在网络架构层面提供策略实施的机制保障成为每个自治域网络的首要目标。本课题面向分布式管理的多自治域环境，围绕缓存管理的自主性、策略设置的灵活性、策略实施的简单性，研究域间的缓存决策策略、通告策略和请求转发策略，提出缓存策略实施所需的信息表示、存储、传输和获取机制。

Pervasive in-network caching is the foundation of many Information-Centric Networking(ICN) architectures. Presently, most of pervasive caching researches focus on improving the performance of a single cache network by cache coordination, regardless of the need for autonomous network management, which is, however, crucial to the commercialization and globalization of ICN. When ICN is extended to the environment with multiple autonomous systems, in-network cache can not only be used to reduce the network provider's own traffic, but also be as a bargaining chip in the process of business relationship negotiation. Consequently, devising appropriate cache policies to realize its business contracts with other networks, and providing mechanisms to enable the policy enforcement becomes the primary goal of each autonomous network. Faced with the multiple autonomous systems that are managed in a distributed manner, this project centers around autonomous cache resource management, flexible cache policy setting and simple cache policy enforcement. This project will study the cache decision policy, content advertisement policy, request forwarding policy between different autonomous networks. Also, this project will design the meta-information representation, storage， transmission and retrieval mechanisms for fulfilling those policies.