Cache Pre-fetching and Replacement strategies for location dependent data in mobile environments

In this paper, new strategies for prefetching and cache replacement are proposed. The proposed pre-fetching algorithm considers the geographical and semantic adjacency between queried items. The strategy is to pre-fetch the most probable secondary service as a by-product of the execution of a query to a primary service. Association rule mining is used for determining the secondary item that should be pre-fetched. The latter part of the paper deals with a cache replacement algorithm PAID+ (Probability Area Inverse Distance +) which takes into account the access probability and distance between the mobile client and the service item which suits a pre-fetching environment. The experimental evaluations using synthetic datasets show the cache pre-fetch and replacement policy PAID+ are effective in improving the system performance in terms of the cache hit ratio of mobile clients.