Improving Packet Caching Scalability Through the Concept of an Explicit End of Data Marker

The web has witnessed an explosion of dynamic content generation to provide web users with an interactive and personalized experience. While traditional web caching techniques work well when redundancy occurs on an object-level basis (page, image, etc.), the use of dynamic content presents unique challenges. Although past work has addressed mechanisms for detecting redundancy despite dynamic content, the scalability of such techniques is limited. In this thesis, an effective and highly scalable approach, Explicit End of Data (EEOD) is presented, which allows the content designer to easily signal boundaries between cacheable and non-cacheable content. EEOD provides application-to-stack mechanisms to guarantee separation of packets with the end goal of simplifying packet-level caching mechanisms. Most importantly, EEOD does not require client-side modifications and can function in a variety of server-sideetwork deployment modes. Additionally, experimental studies are presented, showing EEOD offers 25% and 30% relative improvements in terms of bandwidth efficiency and retrieval time over current approaches in the literature.