Data Modelling

What

Data Modelling is a technique for helping Business Staff to identify and document the data that will be required for an Application System.

Data Modelling can occur at both the logical and the physical level as shown by the following diagrams:

The Logical Data Model does not portray how the data is actually stored in a data base. It shows "What" and not "How".

Entities are people, places or things that we are interested in tracking.

Attributes are characteristics of the people, places or things.

Attributes In Entities are sometimes known as "Entity Relationship Diagram (ERD)".

Relationships show linkages between the entities.

Columns are sometimes known as "Fields".

Tables are sometimes known as

Physical Deployment Architecture includes Client/Server, N-Tier, Web-Based, Firewalls, etc.

Physical Data Base includes disk spindles, controllers, etc.

Why

Data Modelling is performed to:

1. Ensure that the right data is included in the application system.
   
   
2. A good Logical Data Model acts as the foundation for a good application system.

Isn't Data Modelling just an IT function? Why do we Business Staff need to be involved?

1. Because it elicits very specific and detailed Business Rules that are extremely important to know about the required Application System to facilitate the Acquisition and/or Development of a flexible solution.
   
   
2. Only Business Staff know these Business Rules.

When and Where

This technique is used during Enterprise Analysis, Business Area Analysis, System Definition, Design, Development as well as the Maintenance and Support project lifecycle phases.

How

Business Staff determine the right data in the Logical Data Model.

The results of Data Modelling are documented in a Logical Data Model.

See "Joint Application Design (JAD) Workshop Facilitator's Guide" for details on how to perform Data Modelling.

After identifying the first Type or Classification entity, model it in detail. For all subsequent Type/Classification Entities, just have a list on a separate Static Magic sheet with examples of each of the Types.

Also See

Data Modelling is one of the significant Threads that exist throughout PMMentor (PMM). To print a "Handout Pack" on this Tread, create and print a title page based on that shown under the "Methodology Treads" topic and then return to this topic by clicking on the "Back" button, click on the "Print" button (to print this topic to act as a Table of Contents) and then click on each of the following topics in turn. When the topic is displayed, click on the "Print" button, then click on the "Back" button to return to this topic and choose the next topic to print.

Application Architecture

Attribute Naming Standards

Breadth Then Depth

Classifying Data

Common Processes and Cycle Times

Convert Data

Converting Functional Data Model To Physical Data Model

Converting Logical Data Model To Functional Data Model

Data Architecture

Data As A Corporate Resource

Data Characteristics

Data Cleanup

Data Cleanup Team

Data Is More Stable Than Functions

Data Mart

Data Model Implied Functions

Data Model Standards

Data Ownership Versus Stewardship

Data Storage Systems

Data Warehouse

Decompose Compound Fields

Develop Functions/Data Matrix

Develop Logical Data Model

Entity Naming Standards

Entity Relationship Diagramming (ERD)

Existing System Cross-Check

Extended Relational Analysis (ERA)

Function Analysis

Functions/Data Matrix

Generic Data Modelling

Generic Data Subjects

Introduction to Data Modelling

IT Help Desk

IT Information Centre

Joint Application Design (JAD) Workshop Facilitator's Guide

Keys

Logical Data Model

Meta Data

Miscellaneous Data Modelling Issues

Naming Standards For Entity-to-Entity Relationships

Non-Significant Digits In Primary Key Attributes

Normalisation

Referential Integrity

Requirements Management

Standard Entity, Attribute, Process, Procedure, Function and Task Abbreviations

Sub-Typing

Takeup Files