Modified predicted region based cache replacement policy for location-dependent data in mobile environment

摘要

Context-aware information includes time, location, device identity and nearby devices to a given entity. Location-aware information services are one of the important categories of context-aware computing. Location services can be characterized as applications that coordinate a cell phone's area or position with other data in order to give theenhanced value of services to the client at the right time and in the right place. As Mobile client can have limited information in their cache, it needs to find out the data objects that should be replaced when there is no space in client's cache. Earlier cache replacement policies solely based on the temporal features of client's access pattern. This paper proposes a Modified Predicted Region Based Cache Replacement Policy (MPRRP), which considers temporal and spatial properties both for estimating the replacement cost. To ensure efficient cache utilization, MPRRP considers the important factors for data item's cost computation such as client access probability, query rate, update rate, client location and moving direction of mobile clients when employing cache replacement in order to improve the effectiveness of previous cache replacement policy.