Relational Database in RDBMS

    Why use a Relational Database?
•    The purpose of a database is to organize data so it can be easily selected, sorted, managed, and updated.
•     By using a relational database, each piece of information only needs to be stored once.
•    The benefit is that if a piece of data is updated it needs to be updated in only one
place and the update will then be recognized throughout the database.
•    Otherwise, several updates would need to be made in several different places, which can be cumbersome and lead to a messy situation if not done accurately.
•    Example: An update for a student’s address is made in the Student table and
is now recognized by all related tables.

    What We Want From a Relational Design
Good relational database designs share very similar properties to good software designs:
•    Minimal to zero redundancy. Information is stored in one place, without   autonomous copies. Thus, one change affects one clear section of the database, and no changes need to be applied more than once.
•    Let the system work for you. Relational databases implement clear, unambiguous rules in areas such as the behavior and interaction of keys (primary, foreign), the acceptability of null values, and the exact behaviour of relational operations.
•     A good design takes full advantage of these features, allowing the computer to do what it does best, with minimal additional maintenance.
•    Accurate reflection of the domain. In the end, a database’s usefulness is determined by how closely it matches the needs and rules of its application domain.
•    Relationship cardinalities must reflect their real world counterpart; impossible situations in the real world must be similarly impossible (or at least sufficiently improbable) in the database interaction; the most frequently-performed activities in the real world should be the most natural and efficient ones in the database.
   Normalization:-
•    Normalization is a process of decomposing a set of relations with anomalies to produce smaller and well-structure relations that contain minimum or no redundancy.
•    It is a formal process of deciding which attributes should be grouped together in a relation.
•    Normalization provides the designer with a systematic and scientific process of grouping of attributes in a relation.
•    Therefore, the process of normalization can be defined as a procedure of successive reduction of a given collection of relational schemes based on their FDs and primary keys to achieve some desirable form of minimised redundancy, minimised insertion, minimised deletion, and update anomalies.
•    Normalization serves primarily as a tool for validating and improving the logical database design, so that the logical design satisfies certain constraints and avoids unnecessary duplication of data.
•    First goal – eliminate redundant data.
For ex :- storing the same data in more than one table.
•    Second goal-  ensure data dependencies make sense.
For ex :- only storing related data in a table.
•    Normalization is the process if analysing the given relational schema based on its functional dependencies and keys to achieve the desirable properties of:

o    Minimizing redundancy.
o    Minimizing the insertion, deletion, and updating anomalies.
o    Minimize data storage.
o    Unsatisfactory relation schema that do not meet a given normal form test are decomposed into smaller relational schemes that meet the test and hence possess the desired properties.
o    Key concepts in normalization are functional dependency and keys.