Multiresolution Amalgamation: Dynamic Spatial Data Cube

Aggregating spatial objects is a necessary step in generating spatial data cubes to support roll-up/drill-down operations. Current approaches face performance bottleneck issues when attempting to dynamically aggregate geometries for a large set of spatial data. We observe that changing the resolution of a region is reflective of the fact that the precision of spatial data can be changed to certain extent without compromising its usefulness. Moreover most spatial datasets are stored at much higher resolutions than are necessary for some applications. The existing approaches, which aggregate objects at a base resolution, often results in a processing bottleneck due to extraneous I/O. In this paper, we develop a new aggregation methodology that can significantly reduce retrieval (I/O) costs and improve overall performance by utilising multiresolution data storage and retrieval techniques. Topological inconsistencies that may arise during resolution change, which are not handled by current amalgamation techniques, are identified. By factoring these issues into the amalgamation query processing, the retrieval loads can be further reduced with guaranteed topological correctness. Experimental results illustrate significant savings in data retrieval and overall processing time of dynamic aggregation