Course Overview

This course provides a solid basis from which one can successfully implement relational database, multidimensional data warehouse, and business intelligence (BI) technologies. The principal objective of this initial course is to convey a practical and common-sense guide to the theory and concepts of data modeling. Using these sophisticated techniques, one can create an elegant logical design of a database. Within this course we discuss not only the premier modeling theories from the best industry experts but also present the practical and real-world experience of the past 20 years of Sideris data design practitioners.

The methodologies discussed are applicable to any relational database environment, including IBM DB2, the Oracle database, Microsoft SQL Server, and the open-source MySQL and PostgreSQL databases, as well as other RDBMS platforms. They are also applicable to other database technologies, such as object databases and legacy IMS and IDMS databases. Finally, while we use the free Oracle SQL Developer Data Modeler product as a demonstration modeling tool, one can complete the exercises of this course and apply the techniques learned using any other popular data model diagramming tool, such as IBM InfoSphere Data Architect, CA ERwin Data Modeler, Embarcadero ER/Studio, and others.

What you’ll learn

• A review of model-based design, including process modeling, physical data modeling, and other modeling techniques that relate to logical data modeling.
• A comparison of data modeling concepts and theories, including top-down data modeling, bottom-up data modeling, data normalization, object-oriented modeling, and semantic modeling.
• Hints, tips, and guidelines for identifying entities, attributes, and relationships that should appear within a data model.
• Review the popular commercial data modeling tools commonly in use today.
• The benefits of building a conceptual data model in advance of the logical model.
• Learn to find and fix well-known mistakes that can exist in relationship definitions, find missing attributes, and correct erroneous attribute definitions.
• Review a recommended strategy for unique identifiers.
• Using semantic modeling constructs and techniques such as supertypes, subtypes, generalization, specialization, constraints, lattices, and arcs.
• Using object-oriented modeling techniques such as domains, attribute classes, extended types, and abstraction of attributes.
• Time dependency and state dependency within a data model.
• Explore classic structures and modeling patterns, including many-to-many recursion.
• Steps and available options for engineering a physical data model from a logical model.
• Reverse engineering and forward engineering of a physical data model into an implementation relational database.