Using constraints for distributed query processing

Constraints are combined with join operations to minimize the amount of data transmission required for distributed query processing. We identify and exploit two important concepts: reducers and abstractors. A constraint can serve as a reducer if it is a sufficient and/or necessary condition for a query. Through replacing a distributed query by a reducer, the query is converted to be processed only at the local site. This converted query, however, requires “compensation” if the constraint is not a necessarily sufficient condition of the query. Although the compensation of the converted query is typically of database tuples, it can also be of the form of abstraction. This paper also shows that a constraint serves as an abstractor to produce a cooperative answer