Adding time dimension to relational model and extending relational algebra (Q1094156)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Adding time dimension to relational model and extending relational algebra |
scientific article; zbMATH DE number 4024834
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Adding time dimension to relational model and extending relational algebra |
scientific article; zbMATH DE number 4024834 |
Statements
Adding time dimension to relational model and extending relational algebra (English)
0 references
1986
0 references
A new method of incorporating time information into database relations is proposed. While earlier approaches attached time-instant or -interval annotations to tuples, this method (optionally) attaches begin and end times to individual attribute values. The model also allows attribute values that are sets of basic values or time-annotated basic values. The advantage over other approaches is that time-series data on an entity can be collected into a single tuple, rather than spread over multiple tuples, where key and interval ordering considerations complicate extracting time-series data on a single entity. The presence of set- valued attributes means relations in the model are not in first normal form. However, the model here is not as general as other non-first normal models, since nested sets are prohibited. Possibly a simpler system and exposition would have resulted if set constructs were independent of time-attachment, and nesting of sets were allowed. The bulk of the paper is definitions of operations in a relational algebra over time-annotated relations. There are analogs of standard algebraic operators (select, project, join, Cartesian product, union, intersection, difference) that deal with time intervals and set-valued attributes. Nest and unnest operations convert between a set-valued attribute and a collection of tuples with simple attribute values. Other operations convert time annotations to independent attributes and vice- versa. The algebra also contains aggregation operations, and operations to split and combine time intervals. The paper concludes with identities on expressions over the algebra.
0 references
relational database theory
0 references
historical database
0 references
non-first normal form
0 references
time information
0 references
relational algebra
0 references