New Database Architecture for Smart Query Handler of Spatial Database

Abstract

A spatial database system is a database system with additional capabilities for handling spatial data. It also supports spatial data types in its implementation, providing spatial indexing and efficient algorithms for spatial join. The retrieval of data values from a spatial database involves searching through the huge repository of data, involving huge cost. Thus query optimization on spatial database takes more time as compared to RDBMS. The current state of art shows that during the execution of a query in a spatial database management system (SDBMS), the query optimizer creates all possible query evaluation plans. All plans are equivalent in term of their final output but vary in their execution cost, the amount of time to run. Once the data is retrieved the query and its plans are deleted from memory to free the space for future usage. This is repeated for the next query even if the query is already executed. This leads to increased storage overhead and execution time. In this paper, a new database architecture is proposed, which uses a buffer based query optimization technique for faster data retrieval.