Abstract:

In the named data networking (NDN), routers have the capacity of in-network cache, which greatly improve the efficiency of data distribution and retrieval in the network. However, because of the limited cache capacity in routers, how to design an effective caching strategy is still a grave challenge. To solve the problem, we present a caching decision and replacement strategy  based on dynamic content popularity (DPDR). It fully considers content popularity and caching capacity and utilizes an additive increase multiplicative decrease (AIMD) algorithm to dynamically adjust the popularity threshold to store the arriving data whose popularity exceeds the popularity threshold in the cache space. In addition, we also propose a caching replacement algorithm, which takes the historical information of content popularity and the last request time into account to remove the contents with lowest replacement value from the cache store (CS). Simulation results show that compared with other schemes, the DPDR strategy can effectively improve cache hit rate, reduce average cache hit distance and decrease network throughput.

Key words: named data networking (NDN), caching access policy, caching replacement policy, content popularity, AIMD