A data model
The data model is used to represent real facts of the application. An application may contain many facts however one has to focus only on important facts ignoring the others. For example in case of student details application, student name, roll no, address have to be described however student may also contain other facts like fathers name etc. which may not be relevant. Some of the salient features that model must may have are listed.
- Data model mainly describes the data, which gets stored and processed in a given situation.
- A data model may describe data and various levels and description may be at logical / physical levels or from the point of user.
- A data model proposes a set of concepts for description of the nature of data and inter-relationships between them along with the syntax.
- A model should have as minimum concepts, which are close to real world so that user can understand the model and verify.
- The model should provide primitives by which meaning of data can be captured. The meaning contains type of value data items take their inter-relation to higher level entities and the correctness requirement for them.
Relational Database Model
Insertion anomalies and redundant data are problems associated with an early database model known as a hierarchical table (parent-child table). Network database (owner-member table) models were problematic as well. These two models led to the development of the relational database model.
The relational model for database management is a database model based on first –order predicate logic (mathematical theories applied by Dr. E. F. Codd). A database model organized in terms of relational model is a relational database model (RDM).
In a RDM, data are stored in a relation or table (those terms may be used interchangeably.) Each table contains rows or records, (also called tuples), and columns which represent attributes or fields. Each record or row is represented by a unique field known as the Primary key. The categories of relationships in a RDM are one-to-one, one-to-many, and many-to-many. A many-to-many relationship must be broken down into numerous one-to-many relationships. If a pair of tables share a relationship, data can be retrieved based on matching values of a shared field between the tables. Data is retrieved by specifying fields and tables using a standard query language known as Structured Query Language (SQL). Most DBMSs (Database Managements Systems) use SQL to build, modify, maintain and manipulate databases. Thorough knowledge of SQL isn’t always necessary since most DMBSs use a graphical interface to generate SQL statements and retrieve data. It is good, however, to have basic knowledge of SQL.