Data Models: Importance and Building Blocks

Importance of data models

• Data models offer a relatively simple representation of real-world data structures, often using graphical formats.
• Database design tools, such as Visio, can produce useful drawings and automate much of the design work.
• Using a database saves costs, and reduces development time and potential errors.
• Data models improve business understanding and enhance communication among project stakeholders.

Basic building blocks of data models

• An entity represents a real-world object distinguishable from others, such as a specific person, place, thing, or event.
• Attributes are the characteristics that describe entities.
• Relationship is an association among entities
• Relationship examples are: One to many (1:) like one lecturer to many students, many to many (: *) like many students studying many courses, one to one (1:1) like one department having one HOD
• Constraint is a restriction placed on data that ensures its integrity.
  • For example, a student's DP percentage must be between 0 and 100, or each class must have only one teacher per subject.

Business rules

• Business rules provide brief, precise and clear descriptions of an organization's policies, procedures, or principles.
• Business rules applies to organizations that store and use data to generate information.
• They need to be documented in writing and kept up to date
• Often, business rules may originate from outside the organization.
• Business rules should be easily understood and widely shared
• Business rules should describe data characteristics as viewed by the company.
• Sources for business rules include company managers, policy makers, department managers, written documentation (procedures, standards, operations manuals), and direct interviews with end users.
• Nouns generally translate into entities (e.g., Customer, Invoice, Course, Classroom), while verbs represent relationships among entities (e.g., Purchase, Pay, Generate invoice, Attend course).
• Relationships are bi-directional
  • For example, one customer can generate many invoices, and one invoice is generated by only one customer.

Evolution of Data Models

• Data models evolution started with Hierarchical models
• There was Network models
• Relational models emerged after network models
• Entity relationship models were next
• Object oriented (OO) models were created after entity relationship models.
• EF Codd developed the relational model in 1970
  • Codd was a mathematician, computer scientist, and a Royal Air Force pilot during World War II
  • In 1948, he relocated to New York to work for IBM as a mathematical programmer, and later worked for IBM in California until the 1980s
  • Codd received the Turing Award in 1981
  • IBM initially declined to implement Codd's relational model to protect revenue from IMS/DB
  • Codd demonstrated the potential of his model to IBM customers, who then pressured IBM to include a System R subproject in its Future Systems project.
  • A language called SEQUEL was developed, later named SQL

The Relational Model

• "A Relational Model of Data for Large Shared Data Banks" was the paper published in 1970
• It was considered ingenious but impractical in 1970
• The Relational Model is conceptually simple
• In 1970, computers lacked the power to implement the relational model
• Microcomputers today can run sophisticated relational database software
• The Relational Model is implemented with a Relational Database Management System (RDBMS).
• The most important advantage of the RDBMS is its simplifies the complexities of the relational model for the user.
• A table (relation) represents a matrix of row/column intersections.
• Tables are related by sharing a common entity characteristic.
• A table is purely a logical structure.
  • The data's physical storage is not a concern for users or designers.
  • "tables of rows and columns" are like spreadsheets.
• A relational diagram is a representation of the relational database's entities, their attributes, and the relationships between those entities.
• The popularity of the relational model is due to its powerful and flexible query language.
• Structured Query Language (SQL) allows users to specify required actions without detailing how to perform them.
• SQL-based relational database applications include a user interface, a set of tables in the database, and an SQL.

Degrees of Data Abstraction

• Data abstraction is reducing a portion of data for a simple presentation of content
• Abstraction is the process of removing features to reduce the set of necessary features
• Data abstraction is a way of classifying data models
• There are data abstraction levels: The External Model, Conceptual Model, the Internal Model and the Physical Model
• American National Standards Institute (ANSI) Standards Planning and Requirements Committee (SPARC) defined a framework for data abstraction (1970s)
  • The External Model is the end users’ view of the data environment
  • Usage of business rules with external model is highest level of abstraction
  • Easy identification of data for each unit's operations by using advantages
  • The External Model facilitates designer’s job by providing feedback about the model’s adequacy
  • The External Model's creation ensures security constraints in database design and simplifies program development
  • The Conceptual Model Represents global view of the entire database and a Representation of data as viewed by the entire organization
    • Basis for identification and high-level description of main data objects, avoiding details, Most widely used conceptual model is the entity relationship (ER) model
  • Provides view of the data environment, independent of software and hardware
  • It's the representation that the database is "seen" by the DBMS.
  • The Internal Model maps the conceptual model to the DBMS
    • Internal schema depicts a specific representation
  • Physical Model is describes the way data are saved on disks or tapes
    • It's software and hardware dependent
    • Requires designers have detailed knowledge of used implementations