A future location-prediction replacement strategy for mobile computing environments

Abstract

While the traditional data shipping systems use physical information to organize data, a newer data caching technique maintains both associated answers (data items) of previous queries and their meaning (purpose). This form of information description is called semantic description, and it makes it possible to reason about and derive knowledge from the given description. In a distributed heterogeneous environment, such as the mobile environment, the semantic data caching paradigm is an important technique for improving the performance of wireless data dissemination systems. Data caching performance depends heavily on the replacement policy being used for the cache management. However, future location ambiguity, limited client resources and frequent client disconnections make cache management a challenge. This paper proposes a future location-aware cache replacement policy to manage a semantic cache. The proposed replacement policy uses the validity of the data fetched from the server and the neighboring locations to decide which of the cache entries is less likely to be needed in the future, and therefore, a good candidate to be evicted when cache space is needed. The replacement policy's preference for cached items to be evicted is modelled as a general function that declines with distance between the cell the candidate for replacement's cached item is bounded to and the query's bound cell. For better efficiency the overall replacement granularity is dynamically achieved along three levels: ring, cell and data item. Simulation study of the proposed approach shows that it outperforms both the least recently used (LRU) and furthest away replacement (FAR) schemes, where only temporal locality is considered. Moreover, the proposed scheme is easier to implement than the industry standard policies