Network Distance Based Cache Replacement Policy for Location-Dependent Data in Mobile Environment

Abstract

Data caching in mobile clients is an important technique to enhance data availability and to improve data access time. Due to cache size limitations, cache replacement policies are used to find a suitable subset of items for eviction from the cache. The existing policies focus on Euclidean space and consider Euclidean distance as a spatial parameter for eviction. However, in practice the position and movement of objects are constrained to spatial networks where the important distance measure is the network distance. In this paper we propose a cache replacement policy, which considers the network density, network distance and probability of access as important factors for eviction. A series of simulation experiments have been conducted to evaluate the performance of the policy. Results indicate that the proposed cache replacement scheme performs significantly better than the existing policies FAR and PAID.