Design and Implementation of an Object-Oriented SQL Query Processor with an Optimizer

A SQL-like object-oriented query language is implemented on Exodus Storage Manager with an original query optimization strategy based on the object-oriented features of the language. For this purpose the selectivity of a path expression, cost of forward and backward traversals are calculated, and join sizes are estimated. An original strategy is developed for choosing the order of selections and path traversals. The optimization algorithm is designed in such a way that it generates the query execution plan in one pass. This is accomplished through combining the information collected during parsing with the cost estimations. In other words, the algebraic manipulations and transformations are performed by considering the effect of the cost analysis on them. It is clear that in this way the query rewriting becomes more intelligent since the algebraic transformations are performed with apriori information on how well they will perform. 1. Introduction As a part of the MOODS (METU Object-..