Multi step Query Processing
Filter/refinement query processing
A query on a database reports those objects which fulfill a given query predicate. A query processor has to evaluate the query predicate for each object in the database which is a candidate for the result set. Multi-step query processing (filter/refinement query processing) is a technique to speed up queries specifying query predicates that are complex and costly to evaluate. The idea is to save the costs of the evaluation of the complex query predicate by reducing the candidate set for which the query predicate has to be evaluated applying one or more filter steps. The aim of each filter step is to identify as many true hits (objects that truly fulfill the complex query predicate) and as many true drops (objects that truly do not fulfill the query predicate) as possible by applying a less costly query predicate. The remaining candidates that are not pruned as drops or reported as hits in one of the filter steps need to be tested...
- 4.Korn F, Sidiropoulos N, Faloutsos C, Siegel E, Protopapas Z. Fast nearest neighbor search in medical image databases. In: Proceedings of the 22nd International Conference on Very Large Data Bases; 1996.p. 215–26.Google Scholar
- 5.Kriegel H-P, Kröger P, Kunath P, Renz M. Generalizing the optimality of multi-step k-nearest neighbor query processing. In: Proceedings of the 10th International Symposium on Advances in Spatial and Temporal Databases; 2007. p. 75–9.Google Scholar