Below is a well-structured, detailed, and exam-oriented explanation of Various DBMS Data Models.
This covers all types, including classical and modern models — perfect for 5, 10, or 15-mark answers.
DBMS DATA MODELS (Full Explanation)
A Data Model is a way to describe, structure, and organize data in a database.
It defines:
✔ How data is stored
✔ How data is connected
✔ How data can be manipulated
✔ Constraints on the data
A DBMS uses various data models for designing and managing databases.
TYPES OF DATA MODELS
DBMS data models are broadly divided into:
- Object-Based Data Models
- Record-Based Data Models
- Physical Data Models
- Conceptual / High-Level Data Models
- Recent & Emerging Data Models (Modern Databases)
Let’s discuss each in detail.
1. OBJECT-BASED DATA MODELS
These models use real-world objects (entities) and their relationships.
They represent data in a high-level, conceptual way.
(A) Entity–Relationship (ER) Model
- The most commonly used conceptual model
- Represents data as:
- Entities (Student, Teacher, Product)
- Attributes (Name, Age, Price)
- Relationships (Enrolled, Buys, Manages)
Features:
- ER diagrams
- Cardinality (1:1, 1:M, M:N)
- Specialization, Generalization
- Weak and strong entities
Used for database design (conceptual schema).
(B) Object-Oriented Data Model
- Based on objects, classes, inheritance, encapsulation
- Data + methods stored together
- Supports complex data types (images, video, multimedia)
Used in:
- CAD/CAM
- Real-time systems
- Multimedia applications
(C) Object-Relational Data Model
- A hybrid between relational and object-oriented models
- Allows user-defined types (UDTs)
- Allows complex objects, arrays, nested tables
- Used in: PostgreSQL, Oracle, DB2
2. RECORD-BASED DATA MODELS
Represent data in fixed-format records (rows).
These are used by traditional DBMS (RDBMS).
(A) Relational Data Model (Most Important)
- Proposed by E.F. Codd
- Data stored in tables (relations)
- Rows = tuples, Columns = attributes
- Uses primary key, foreign key, and constraints
Features:
- Simple, flexible, and mathematically sound
- SQL used for data manipulation
- Ensures data integrity via constraints
Examples:
MySQL, PostgreSQL, SQL Server, Oracle
(B) Hierarchical Data Model
- Data organized in a tree structure
- Parent–child relationship (1-to-many)
- Each child has only one parent
- Navigational access using pointers
Example:
Organization
|
Employee
|
Dependent
Used in early systems and still used in IBM IMS.
(C) Network Data Model
- Generalized form of hierarchical model
- Supports many-to-many relationships
- Data represented as a graph
- Uses set and record types
More flexible than hierarchical model.
Used in:
- CODASYL DBMS
- Early business applications
3. PHYSICAL DATA MODELS
Describe how data is actually stored on disk.
Includes:
- File organization (heap, sorted, hashed)
- Index structures (B-Tree, Hash Index)
- Record formats
- Page/block structure
- Access paths
- Storage strategies
Used by DBMS developers to optimize performance.
4. CONCEPTUAL / HIGH-LEVEL DATA MODELS
These models present data in a user-friendly, abstract form.
Includes:
- ER Model
- UML Class Diagrams
- Semantic Data Models
- Enterprise Models
Used in:
- Database design phase
- Communication with non-technical users
5. MODERN / EMERGING DATA MODELS (NO SQL & BIG DATA)
New databases support unstructured & semi-structured data.
(A) Document Data Model
- Data stored as documents (JSON, BSON, XML)
- Schema-less
- Supports nested objects
Examples:
- MongoDB
- CouchDB
(B) Key–Value Data Model
- Simple model with two elements:
- Key (unique)
- Value (any data)
Used for:
- Caching
- Real-time applications
Examples:
- Redis
- DynamoDB
(C) Column-Family Data Model
- Data stored in columns instead of rows
- Highly scalable
- Used in Big Data systems
Examples:
- Cassandra
- HBase
(D) Graph Data Model
- Data represented as nodes and edges
- Used for highly interconnected data
Examples:
- Neo4j
- Amazon Neptune
- Facebook social graph
Used in:
- Social networks
- Recommendation systems
- Fraud detection
SUMMARY TABLE (Perfect for Exams)
| Type | Data Model | Description | Examples |
|---|---|---|---|
| Object-Based | ER Model | Entities, attributes, relationships | Database design |
| Object-Based | Object-Oriented | Objects, classes, inheritance | OODBMS |
| Record-Based | Relational | Tables, rows, columns | MySQL, Oracle |
| Record-Based | Hierarchical | Tree structure | IBM IMS |
| Record-Based | Network | Graph structure | CODASYL DB |
| Physical | Physical Model | Storage-level details | DBMS internals |
| Modern | Document | JSON documents | MongoDB |
| Modern | Key–Value | Key-value pairs | Redis |
| Modern | Column Family | Column-based storage | Cassandra |
| Modern | Graph | Nodes and edges | Neo4j |
Perfect 5-Mark Answer
DBMS data models describe how data is structured, stored, and manipulated in a database. The major data models are:
- Relational Model – stores data in tables; uses primary & foreign keys; most widely used.
- Hierarchical Model – organizes data in a tree form with parent–child relationships.
- Network Model – uses graph structure supporting many-to-many relationships.
- Entity–Relationship Model – high-level conceptual model using ER diagrams.
- Object-Oriented Model – stores data as objects with methods.
- Modern Models – include document, key-value, column-family, and graph models used in NoSQL databases.