A Uniform Data Model for Relational Data and Meta-Data Query Processing

We introduce the uniform data model to reason about relational data and meta-data query processing. The principal property of the uniform data model is that both meta-data and data are organized in uniformly shaped information objects. This uniform treatment of meta-data and data allows for the design of query languages that are independent of the meta-data/data separation. We present two query languages that are designed in this manner: the uniform calculus and the uniform algebra, which are a logical and algebraic query languages, respectively, for the uniform data model. We establish that the uniform calculus has at least the expressive power of the uniform algebra and prove that, besides providing meta-data query processing capabilities, these languages can e ciently simulate conventional relational query languages. We also give upper-bounds on the expressive power of these languages and since these languages have higher data complexity than the relational calculus and algebra, we introduce sublanguages that have AC data complexity.