Observational Properties of Databases with Object Identity

Anthony Kosky

Technical Report MS-CIS-95-20,
University of Pennsylvania, 1995.

In this paper we will study the problem of distinguishing between database instances and values in models which incorporate object-identities and recursive data-structures. We will show that the notion of observational distinguishability is intricately linked to the languages available for querying a database. In particular we will show that, given a simple query language incorporating a test for equality of object-identities, database instances are indistinguishable iff they are isomorphic, and that, in a language without any operators on object-identities, database instances are indistinguishable iff a bisimilarity relation holds between them. Further, such a bisimulation relation may be computed on values, but doing so requires the ability to recurse over all the object-identities in an instance.

We will then go on to look at systems of keys, and show that these give rise to observational distinguishability relations which lie between these two extremes. We show that a system of keys satisfying certain restrictions provides us with an efficient means of comparing values, while avoiding the need to compare object identities directly.

See here for the paper.


Back Back to DB Group Homepage

sharker@saul.cis.upenn.edu