Database Models and Big Data Concepts Quiz

Data Models

• Data models are relatively simple representations of complex real-world data structures.
• They are often graphical to facilitate understanding for designers, programmers and end-users.
• Effective communication between these roles using data models is critical.
• Data models are necessary for organizing data for various users
• Data Models use:
  • Entities (person, place, thing, event)
  • Attributes (characteristics of an entity)
  • Relationships (associations between entities)
  • Constraints (restrictions on data to ensure integrity)
  • Business rules are useful when understanding entities, attributes, relationships and constraints

Data Modeling Building Blocks

• Entity: Something about which data are collected and stored. Examples: Customer, Student, Patient, Block, Room, Book, CD, Furniture, Wedding, Musical Concert. It represents a specific type of object.
• Attribute: A characteristic of an entity. Examples: CUST_ID, CUST_ADDRESS. They describe the property of an entity.
• Relationship: An association among entities. Examples of types are 1-to-many (1:M), many-to-many (M:N), and one-to-one (1:1). Describe the connection between different instances (entities).
• Constraint: Restrictions placed on the data to ensure integrity. Examples: a student's GPA must be between 0.00 and 4.00, an employee's salary must be between 6,000 and 350,000). Rules that must be followed to maintain data quality.

Business Rules

• Brief, precise, and unambiguous descriptions of policies, procedures or principles within an organization.

• Give descriptions of operations that help create and enforce actions within an organization's environment.

• Examples: A customer may generate many invoices, an invoice is generated by only one customer.

  A training session cannot be scheduled for fewer than 10 or more than 30 employees.

• Sources of business rules: Company managers, Policy makers, Department managers, Written documentation, Procedures, Standards, Operations manuals, Direct interviews with end-users.

• Business rules must be rendered in writing, kept up to date and easily disseminated. They define the characteristics of the data as viewed by the company.

Data Model Evolutions

• A summary of data model generations is included in a table. Example table data includes generation, time, data model, examples and comments. Generations include File system, Hierarchical, Network, Relational, Object-oriented, Object-relational, XML and Hybrid DBMS

Hierarchical Model

• Developed in the 1960s to manage large amounts of complex data for manufacturing projects. It uses an upside-down tree structure and 1:M relationships.
• The primary disadvantage is its complexity, difficulty of managing and structural dependencies. There is also a lack of standards.
• It formed the foundation of current data models.

Network Model

• Created to depict relationships more effectively than the hierarchical model by improving database performance and imposing a database standard,
• It's conceptually similar to the hierarchical model but permits a record to have more than one parent, resolving a limitation of the hierarchical model to contain a single set of 1:M relationships.
• A disadvantage is its lack of ad hoc query capability and difficult maintenance that impacted programmers.

Relational Model

• Developed by Codd in the 1970's, relational model is a table-based design that's popular and conceptually simple.
• Relational model excels by storing data in tables and using a common attribute to relate them. The Relational Model is also suitable for microcomputers.
• One advantage is its ability to hide complexities from the end-users, which makes it easier to manage and maintain. Disadvantages are substantial hardware and system software overhead, the difficulty in conceptual simplicity for untrained users to utilize and may promote information problems

Entity Relationship Model (ERM)

• A graphical tool used for data modeling, introduced in 1976 by Chen. It complements the Relational model by graphically depicting entities and relationships.
• Advantages include conceptual simplicity due to visual representation and effective communication due to its visualization. ERM connects to relational models. Disadvantages include limited data manipulation, restricted relationship constraints and loss of information in crowded displays.

Object-Oriented Model

• Combined data and relationships in a single structure called "object."
• It is conceptually useful for representing real-world entities and providing an inheritance feature to simplify complexity.
• A semantic data model, it acts as a basis for object-oriented database management systems.

NoSQL

• A large scale distributed database system and a new approach to data storage. It's based on structure and unstructured data.
• It's better than the Relational Model for storing datasets with less complexity, particularly Big Data.

Degrees of Data Abstraction

• A framework for classifying data models, often used in database design processes.
• The framework proceeds from a high level of abstraction to an increasingly detailed description. There are generally three degrees of abstraction- External, Conceptual and Internal.
• ANSI/SPARE Committee has a framework. External model is specific views of a user, Conceptual Model views the global database, and Internal schema depicts DBMS model..

Other Models

• Extended Relational Data Model (ERDM)
  • A semantic data model developed in response to increasingly complex applications.
  • It extends the capabilities of the relational model.

Data Models Comparison

• A table presents a comparison of the different data models
  • Each row represents a Data Model
  • Each column represents the category or aspect compared

Emerging Models

• Big data, a new data approach that makes it easier to manage large amounts of data and derive insight from web-generated data. Big-data modeling is useful for massive amounts of varied and high-volume data.