Optimal Caching for Producer Mobility Support in Named Data Networks

Named Data Networks (NDNs) offer a promising paradigm for the future Internet to cope with the growing demand for data. One of the main challenges in NDNs is how to support a seamless operation during mobility. In this paper, we investigate optimal caching for Producer mobility support and propose a scheme (named OpCacheMob) that exploits location predictors and data requests' patterns to cache the data proactively before handover occurs. In essence, OpCacheMob adopts the predicted future Interests, that will be sent to the mobile producers, and caches their data contents ahead. Thus, avoids Interest retransmission or redirection that increase the consumer's delay and decreases the network efficiency during producer's mobility. We provide a mathematical formulation for such caching problem that bounds both the cache update cost and the consumer delay while minimizing the total network overhead due to the change of content availability. OpCacheMob is then implemented in ndnSIM and evaluated against mainstream NDN mobility solutions. We demonstrate how the scheme can be used as a benchmark to measure the performance of other mobility schemes. In addition, a sensitivity analysis is presented to measure the impact of errors on the prediction gain of such solution.