Cooperative caching techniques for named data networks

Abstract

Information Centric Networking (ICN) is a new paradigm of network architecture that makes content a first-class citizen. Named Data Networking (NDN) is a specific implementation of ICN, built on a content-centric communication model to address challenges posed by the current Internet architecture. Unlike the present TCP/IP model, in NDN, communication is no longer based on the location of content; instead, it uses the name of the content itself to fetch the desired data. It is a promising architecture for the future Internet due to its wide range of features, such as content caching in network routers, mobility support, efficient multicast forwarding, scalability, and data authenticity and integrity through per-packet signatures. The in-network caching feature of NDN has drawn significant interest from researchers because it improves content accessibility by storing data closer to end users. Since NDN routers have limited cache space, content is stored temporarily, and when the cache is full, newly arriving content replaces older content. Therefore, designing an effective caching mechanism is essential to enhance user experience and reduce the load on the server. One way to better utilize the available cache space is to bring cooperation among the nodes. In this thesis, we design several cooperative caching techniques to make use of the content available in nearby routers, aiming to improve the cache hit ratio and reduce content access time. Our first contribution in this thesis is eNCache, which is a cooperative content lookup and caching technique. eNCache is designed to efficiently search for content available in the caches of nearby routers outside the transmission path.