CaDaCa: a new caching strategy in NDN using data categorization

Named data networking (NDN) is one of the emerging and promising paradigms of the future internet architecture. An essential feature in NDN is in-network caching, where the contents are decoupled from their physical locations to achieve an effective and consistent content distribution. However, caching capacity in NDN is relatively small and cannot hold all the contents requested by users. Therefore, cache space management becomes critical to address and maintain. In the literature, several solutions have been proposed to properly cache data transferred between servers and users. Most of the proposed approaches are based on name-level request analysis. Unfortunately, processing requests at the name level remains a challenging task because of their unlimited number and rapid volatility over time. In this work, we explore the role of data categorization in enhancing the cache mechanisms in NDN. We present a new caching strategy called CaDaCa (Categorized Data for Caching), where popular requests are categorized enabling in-depth knowledge about users’ behavior, which can be valuable in creating a more powerful caching mechanism. The performance of the proposed approach is evaluated using a real-world data-set extracted from browsing history. Compared to other well-known strategies in the literature, our caching strategy achieves a higher cache hit ratio, a higher hop reduction ratio, and therefore, makes more accurate data caching decisions.