WILLIAM G. SMITH & ASSOCIATES INFORMATION RESOURCE MANAGEMENT SEMINARS AND CONSULTING "CLEANING UP THE WORLD'S DATA MESSES"

OVERVIEW OF DATA MODELING

This informational (not skill-building) seminar presents a complete overview of the three levels/types of data model which are typically used to manage data in a progressive enterprise: Conceptual Data Model, Logical Data Model, and Physical Data Model. The context in which each of these types of models are built is discussed, along with the purpose, form, and content of each type of model. All terms associated with each type of data model are defined; the different types of diagrams and diagram symbols used in each type of data model are presented, and the typical steps followed in building each type of data model are also described. Within each type of model, example diagrams and definitions for each different type of artifact illustrate the typical level of detail and rigor sought for each model. The way that a modeling dictionary/repository would be used to document each type of model, and keep them properly related is shown. The profound value of rigorously modeling data, and preserving data models is clearly communicated.

Due to the amount of material covered, this seminar does not have workshops to build usable skills.

TOPICAL OUTLINE:

• Background Concepts
  • The IRM Environment Defined
  • Traditional "Systems" Approach
  • Fundamental "Systems" Problem
  • Information Resources
  • Shared Data Resource Concept
  • IRM Requires . . .
  • Development Under the IRM Approach
  • IRM Critical Success Factors
  • Data Modeling Defined
  • The Conceptual Data Model in Context
  • Data Model Continuity
  • Time and Data
  • Data Modeling Sessions - Logistics and Success Factors
• Conceptual Data Modeling Overview
  • Purpose, Form and Content of Conceptual Data Model
  • Conceptual Data Modeling Definitions
  • Conceptual Data Modeling Steps - Overview
    • 1. Detecting and Qualifying Candidate Entities
      • Entity Rules
      • Example Entities
    • 2. Diagramming the Entities/Relationships
      • E/R Diagram Symbols
      • Relationship Rules
      • Best Fit Decisions of the Modeler
      • Special E/R Constructs
        • N-ary Relationship
        • Recursive Relationship
        • Subtype/Supertype
        • Characteristic Entity
        • Associative Entity
        • Relationship Roles
        • Role/Contributor Entity
      • Example E/R Diagram
    • 3. Analyzing and Defining E/R States
      • State/Transition Analysis
      • Example State/Transition Diagrams
    • 4. Fully Defining Entities and Relationships
      • Example Entity Definition
      • Rules for Selecting Primary Keys
      • Example Relationship Definition
    • 5. Reviewing E/R Model for Efficacy and Stability
  • Logical Data Modfeling Overview
    • Purpose, Form and Content of Logical Data Model
    • Logical Data Modeling Definitions
    • Alternative Approaches
    • Logical Data Model - Big Picture
    • Logical Model Diagrams
    • Logical Modeling Steps - Overview
      • 1. Identifying Pertinent Transactions/Dataviews for Analysis
      • 2. Analyzing the Transaction/Dataview
      • 3. Standardizing and Defining Required Data Elements
        • Choosing Best Representation
        • Rules for Data Elements
        • Example Data Element Definition
        • Naming of Data Elements
        • Use of Keyword Glossaries
        • Checking the Dictionary for Redundancy
      • 4. Diagramming and Normalizing the Dataview
        • Bubblecharting Symbols
        • Example Bubblechart
        • Bubblecharting the Data for a Transaction/Dataviews
        • Bubblechart/Logical Structure Rules
        • The Normal Forms
          • First Normal Form Examples
        • Example Bubble Chart from Transaction/Dataview Spec.
      • 5. Fully Defining New LDG's, Associations, Entities, Relationships
        • Example Logical Data Group Definition
        • Logical Data Group Naming
      • 6. Verifying the Bubblechart
      • 7. Synthesizing Bubblechart into Composite Logical Model
        • Synthesis Example
      • 8. Reviewing and Stabilizing the Logical Data Model
  • Physical Data Modeling Overview
    • Purpose, Form and Content of Physical Data Model
    • Physical Data Modeling Definitions
    • Pysical Design Issues/Constraints
    • Physical Modeling Steps - Overview
      • 1. Formalizing and Weighting Design Objectives
      • 2. Defining Physical and Technological Environment
      • 3. Laying Out First Cut Physical Design(s)
        • General Logical to Physical Transform (Relational)
        • Physical Data Model Diagram Symbols
        • General Logical to Physical Transform
        • General Foreign Key Rules
        • Example Transforms
        • Short Detour into the World of Distributed Data
          • Distribution Definitions
          • Data Distribution Modes
          • Qualitative Analysis Steps
          • Example Distributed Design
          • Quantitative Analysis
      • 4. Deciding Stored vs. Virtual Derived Data
      • 5. Analyzing and Adjusting for Volume and Growth
      • 6. Analyzing and Adjusting for Security Requirements
      • 7. Analyzing and Adjusting for Transaction Performance
        • Modeling Transaction Data Usage Patterns
        • Performance Prototypes
        • Possible Adjustments to Improve Performance
          • Clustering/Separating
          • Denormalizing and Side Effects
          • Secondary Indices
      • 8. Analyzing and Adjusting for Ease of Use
      • 9. Assessing Design Objectives
      • 10. Finalizing the Physical Model
      • 11. Specifying the Design in DBMS DDL
  • Data Modeling in the Development Scenario
    • Traditional Development Scenario
    • IRM Development Scenario
• Course Summary

DURATION: 5 days

TARGET AUDIENCES: (no recommended maximum number of attendees)

• CIO, IS/IRM Management
• Conceptual, Logical, Physical Data Modelers
• Process Analysts/Modelers
• Development Project Managers
• Business persons who will participate in Data Modeling projects

PREREQUISITES: none