Different types of data models

Data modeling is a process of creating data models for storing in the databases. It is a conceptual as well as a visual representation of various data that are significant and related to a business enterprise. It consists of information, rules, compliances, policies, securities, value, and semantics that are important for a business.

There are three types or stages of data models:

  1. Conceptual Data Model
  2. Logical Data Model
  3. Physical Data Model

Conceptual Data Model

The conceptual data model consists of entities and relationships between entities. The conceptual model provides an overview of the data and is highly abstract; that is the data presented in the conceptual model are general rather than specific. One does not need software to create a conceptual data model as it can be drawn on any piece of paper. One can make diagrams using boxes and lines to represent entities and relationships respectively.  For example; let us take four boxes and name them Time, Product, Sales, and Store, and draw a line linking Time and Product with Sales and sales with the store. Thus, a conceptual model of the data is ready.

The conceptual model consists of the semantics of the data; that is ‘what’ of the data. It does not indulge in descriptions and metadata; rather it shows the relational data between entities. 

Therefore, the conceptual model is highly abstract and consequently easy to understand. They can be easily enhanced as only entities and relationships are visible. One does not require software to create conceptual data models.

Logical Data Model

The next stage is the Logical Data Model. Logical Data Models are an expansion of Conceptual Data Models. It is more descriptive than conceptual models as it includes attributes or characteristics of the entities. The presence of attributes for each entity makes logical models more detailed. The construction of the Logical Data Model consists of Key as well as Non-Key attributes, and a relationship between Primary and Foreign Key attributes. If we have to build from the conceptual model created above for our logical model, it would have attributes such as time, date, month, and year (and more) in the time box. Product id, description, category, price, etc. in the product box. Items sold and sales amount in the sales box and region, store description in the storage box. The sales box will also contain Foreign Key (primary keys in other entities) attributes.

The logical Data Model has more descriptions than Conceptual Data Model. It has user-friendly attribute names and takes more effort to enhance. Logical Data Model requires software programs such as ERWin for construction.

Physical Data Model

Physical Data Models are an expansion of Logical Data Models. It consists of meta-data and has all the information regarding the business. These models can only be designed with the help of software such as ERWin, and the nomenclature of entities and their attributes are changed to tables and columns. These names are database compatible and are not user friendly as only technical people can comprehend it. It is difficult and requires more effort to enhance.   

Comments are closed.

  • Partner links