DBMS Data Models

1. Flat File DBMS

The flat-file or file-management approach was the earliest method for storing data. Each data item was stored sequentially in a large file. To locate a record you typically had to scan from the beginning until a match was found.

Drawbacks: data duplication, poor security, slow retrieval, no support for complex data types, lack of concurrency control, and no standardized access control.

2. Relational DBMS

Proposed by E. F. Codd (1970). The relational model has two core components:

1. A collection of relations (tables) to store data
2. A set of operations (e.g., relational algebra) that produce relations

Why common: relational DBMS are widely used today due to strong data integrity, tabular representation (rows & columns), constraints, normalization to reduce redundancy, support for complex queries (SQL), and multi-user concurrency.

3. Hierarchical DBMS

The hierarchical model organizes data in a tree-like structure (parent → child). It models one-to-many relationships naturally, but repeating child data for multiple parents can cause redundancy.

4. Network DBMS

Introduced in the 1970s (CODASYL), the network model allows many-to-many relationships between records. It reduces some duplication seen in hierarchical models by allowing child records to have multiple parent links.

5. Object-Relational DBMS

Combines relational databases with object-oriented features: support for objects, classes, inheritance, and richer data types. Object-relational DBMSes provide a middle ground between relational and full object-oriented databases and are useful when applications need complex types alongside relational querying.