Efficient cost models for spatial queries using R-trees

Selection and join queries are fundamental operations in Data Base Manageme nt Systems (DBMS). Support for nontraditional data, including spatial objec ts, in an efficient manner is of ongoing interest in database research. Tow ard this goal, access methods and cost models for spatial queries are neces sary tools for spatial query processing and optimization. In this paper, we present analytical models that estimate the cost (in terms of node and dis k accesses) of selection and join queries using R-tree-based structures. Th e proposed formulae need no knowledge of the underlying R-tree structure(s) and are applicable to uniform-like and nonuniform data distributions. In a ddition, experimental results are presented which show the accuracy of the analytical estimations when compared to actual runs on both synthetic and r eal data sets.