Geo Database Management Systems

This course is about managing geo-information in a database management system (DBMS). The geo-DBMS is the central component of the geo-information chain. All information observed and interpreted should be managed in a manner that supports further use, so the Geo-DBMS is the end-point of data collected. Analysis, visualizing and (web-) dissemination of the geo-information start from the geo-DBMS. It is crucial that the content of the geo-DBMS is clear (the information model and its associated semantics) to the users, but also that it is high performant, as the geo-data volumes are often very high and especially in a web-based environment there may be many uses. The course consists of three parts; the first two concern introductions. The main part is devoted to spatial modelling and geo-DBMS. More specifically the three parts cover:

1. Introduction object-oriented information modelling: UML and more specifically class diagram to create conceptual models. Mastering (and applying in design) the key notions such as: Class, Abstract Class, Identifiers, Instances, Attribute, Operation, Association, Generalization, Aggregation, Composition, Multiplicity, Note, Constraint, Package, Stereotype, etc.

2. Introduction relational database management systems: the relational model, SQL DDL (schema definition) and DML (queries). Relational algebra. Some implementation issues (clustering and indexing techniques, such as the b-tree, primary and foreign key constructions supporting referential integrity). In addition a first introduction to the benefits of NoSQL (Not Only SQL) databases.

3. Spatial databases: theoretical aspects of spatial modelling (object and field models), DBMS-GIS architectures (dual, layered, integrated), spatial-temporal data model, Abstract Data Types (ADTs) for geometric primitives, topology model (ISO 19107), temporal aspects, spatial indexing (quad tree, r-tree, field-tree), spatial data clustering (Hilbert, Morton, Cantor, row, row prime, etc.), geo-DBMSs (data modelling, structures, queries, visualisation), (non-)commercial systems (Oracle Spatial, PostGIS, etc.). One or more advanced topics: 3D, Raster, Point cloud data, Simplicial complexes, Geo-OCL, Routing in database, Big data, NoSQL (spatial), Vario-scale, 5D modelling (deeply integrated 3D space, time and scale).