Generating efficient plans for queries using views

We study the problem of generating efficient, equivalent rewritings using v iews to compute the answer to a query. We take the closed-world assumption, in which views are materialized from base relations, rather than views des cribing sources in terms of abstract predicates, as is common when the open -world assumption is used. In the closed-world model, there can be an infin ite number of different rewritings that compute the same answer, yet have q uite different performance. Query optimizers take a logical plan (a rewriti ng of the query) as an input, and generate efficient physical plans to comp ute the answer. Thus our goal is to generate a small subset of the possible logical plans without missing an optimal physical plan. We first consider a cost model that counts the number of subgoals in a phys ical plan, and show a search space that is guaranteed to include an optimal rewriting, if the query has a rewriting in terms of the views. We also dev elop an efficient algorithm for finding rewritings with the minimum number of subgoals. We then consider a cost model that counts the sizes of interme diate relations of a physical plan, without dropping any attributes, and gi ve a search space for finding optimal rewritings. Our final cost model allo ws attributes to be dropped in intermediate relations. We show that, by car eful variable renaming, it is possible to do better than the standard "supp lementary relation" approach, by dropping attributes that the latter approa ch would retain. Experiments show that our algorithm of generating optimal rewritings has good efficiency and scalability.