Replacement Method based on Access Spatiotemporal Locality in a Heterogeneous Distributed Cluster-based Caching System for WebGIS

ABSTRACT

Community user access of a WebGIS is characterized by intensity and popularity. The requested geospatial data have the characteristics of spatial and temporal locality. This paper proposed an expression for the replacement feature by balancing spatial and temporal locality as well as long-term and short-term popularity in tile access to ensure that the replacement process can not only optimize global access but also adapt to the access pattern changes. Then, using the Hash function and linear linked chains to provide cooperative cache management in a heterogeneous cluster-based caching system speeds up the query and replacement process of tiles and improves the performance of the cluster-based cache service. Experimental results revealed that the proposed method obtained a higher cache hit rate and a good average response performance for a heterogeneous distributed cluster-based cache system, providing service to more users and thus increased its service capacity.

Keywords: