We show that some relational queries, which we call quantified queries are not well supported in distributed environments. We give a formal definition of quantified queries, propose a language in which to express said queries and provide a procedure to compute answers in this new language in the context of distributed databases. The proposed language is made up of high-level, declarative operators (called generalised quantifiers), and therefore it can be used in combination with several distributed frameworks. Our approach is designed to be as general as possible; it assumes horizontally partitioned relations, but nothing else, so no data placement or replication is used. We present an implementation and algorithms for the new language, propose some basic optimisations and give experimental results which show that the new approach is indeed quite efficient and scales well.

Optimizing queries in a distributed database is quite difficult. This work proposes defining new generalized quantifiers which operate on sets rather than tuples. These quantifiers would allow for easier optimization in a horizontally distributed database.These operators are scalable with respect to both the number of hosts in the environment and the size of the data used.iv