Recently, we have extended the relational data model to in-corporate linear orderings into data domains [8], which we call the or-dered relational model. We herein formally define Ordered Functional Dependencies (OFDs) and Ordered INclusion Dependencies (OINDs) for the extended model. We show that the conventional sound and com-plete axiom systems for FDs and INDs can be generalised to the cases of OFDs and OINDs. We investigate a subclass of ordered databases called ordered object databases which consists of a set of ordered rela-tions having a distinguished key attribute and enables us to view tuples as linearly ordered objects. An ordered object database possesses two desirable properties concerning OFDs and OINDs, which are useful in ordered database design. First, there is no interaction between OFDs and OINDs. Second, the implication problem for a given set of OINDs, I, whose complexity is polynomial-time in the size of I.
CITATION STYLE
Ng, W. (2000). Using an ordered key attribute in database design. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1873, pp. 193–202). Springer Verlag. https://doi.org/10.1007/3-540-44469-6_18
Mendeley helps you to discover research relevant for your work.