Dynamics and cachability of Web sites: implications for inverted capacity networks

摘要

The traditional Internet access model involves low bandwidth last-mile circuits and high bandwidth backbones. Imagine that in the future the last-mile becomes a high bandwidth service. In such an inverted capacity network content caching in the access network becomes essential to avoid backbone congestion and improve user experience but on the other hand the high access bandwidth also offers opportunities for new caching mechanisms. We focus on the Web as the most important and well-established content service. With respect to caching the question is how much of the Web content is cachable and what is the dynamic behavior? In this paper we analyze the cachability and dynamic behavior of a number of Web sites and the implications for an inverted capacity network. In contrast to previous work we use an active approach for collecting the measurement data to be able to analyze complete Web sites instead of subsets accessed by a specific user group over a certain time period.