| Popis: |
Content Centric Networking (CCN) is a content name-oriented approach to disseminate content to edge gateways/routers. In CCN, a content is cached at routers for a certain time. When the associated deadline is reached, the content is removed to cope with the limited size of content storage. If the content is popular, the previously queried content can be reused for multiple times to save bandwidth capacity. It is, therefore, critical to design an efficient replacement policy to keep popular content as long as possible. Recently, a novel caching strategy, named Most Popular Content (MPC), was proposed for CCN. It considers the high skewness of content popularity and outperforms existing default caching approaches in CCN such as Least Recently Used (LRU) and Least Frequency Used (LFU). However, MPC has some undesirable features, such as slow convergence of hitting rate and unstable hitting rate performance for various cache sizes. In this paper, a new caching policy, dubbed Fine-Grained Popularity-based Caching (FGPC), is proposed to overcome the above-mentioned weak points. Compared to MPC, FGPC always caches coming content when storage is available. Otherwise, it keeps only most popular content. FGPC achieves higher hitting rate and faster convergence speed than MPC. Based on FGPC, we further propose a Dynamic-FGPC (D-FGPC) approach that regularly adjusts the content popularity threshold. D-FGPC exhibits more stability in the hitting rate performance in comparison to FGPC and that is for various cache sizes and content sizes. The performance of both FGPC and D-FGPC caching policies are evaluated using OPNET Modeler. The obtained simulation results show that FGPC and D-FGPC outperform LRU, LFU, and MPC. |
