What is Network Model in database environment

Network Model

A network model was developed to overcome the problems of a hierarchical data model. It modified the hierarchical model by allowing multiple parent-child relationships. These relationships are known assets in the network model. Its structures and language constructs
were defined by CODASYL (Conference on Data Systems Language).

Characteristics of Network Model
The characteristics of a network DBMS are as follows:

1. Data Record Types are represented as Network
In this model, data record types are represented as a network.

2. Each sub-module can have one or more super-modules
Since multiple parent-child relationships are supported, child record types could have more than one parent record type.

3. Represents a set of related records
The sets that support multiple parent-child relationships and the structure of records have to be specified in advance.
 

4. Complex Structure
It supports multiple parent-child relationships that make database structure complex.

5. Relationships are Predefined
Network database implements sets that support multiple parent-child relationships. The sets have to be specified in advance. In the tradeoff between flexibility and performance, a network model is not very flexible to reorganize but has high-performance levels.

6. Difficult to Reorganize
The network database is very difficult to reorganize. The insertion or deletion of the record involves tracing the pointers and changing the appropriate links.

7. Navigation is done by Programmer
The programmer has to write 3GL programs to specify the relationship and direction to navigate the database. Record-by-record navigation is required to access a particular record.

8.3GL Inadequate for Handling Sets
The records in a network model are processed one set at a time. 3GLs handle only one record at a time and hence are inadequate for handling the sets.

9. Query Facility not Available
Network database management systems do not have any query facility. 3GL programs
have to be written specifying the path and the relationship.

Relational Model

Dr. E. F. Codd worked to improve the working of DBMSs to handle large volumes of data. He applied the rules of mathematics to solve the problems of earlier database models. Some important problems were as follows:

1. Data Integrity
2. Data Redundancy

Dr. Codd presented a paper A Relational Model of Data for Large Shared Databanks in June 1970 that contained 12 rules. A DBMS that satisfies these rules is called a full Relational Database Management System (RDBMS). The term relation is also derived from the set theory of mathematics.
In a relational model, data is stored in relations. The relation is another term used for the table. A table in a database has a unique name that identifies its contents. Each table can be called an intersection of rows and columns. An important property of a table is that the rows are unordered. A row cannot be identified by its position in the table. Every table must have a column that uniquely identifies each row in the table.

Relational Database Terminology
Some important terminologies used in the relational database model are as follows:

Relation
In a relational model, data is stored in relations. A relation is another term used for a table.

Tuple

In a relational model, every relation or table consists of many tuples. Tuples are also called records or rows.

Attributes
An attribute is a named column of a relation. Attributes are also called characteristics. The characteristics of the tuple are represented by attributes or fields.

Domain
A domain is a collection of all possible values of one or more attributes. For example, the value in the field "Class" can be the name of any taught class. It is known as the class domain. Similarly, the Registration domain is a collection of all possible Registration numbers.
Degree

The number of attributes is called the degree of that relation.

Cardinality
The number of tuples is called the cardinality of that relation.

Advantages of a Relational Database Model
Some important advantages of a relational database model are as follows:

1. Data Integrity
The relational model allows data integrity from field level to table level to avoid duplication of records. It detects records with missing primary key values at the relationship level to ensure valid relationships between relations.

2. Data Independence
The implementation of the database will not be affected by changes made in the logical design of the database or changes made in the database software.

3. Structural Independence
Structural independence exists when the structure of the database can be changed without affecting DBMS's ability to access the data. The relational database model does not use a navigational data access system. The data access paths are irrelevant to relational database designers, programmers, and end-users. Any change in relational database structure does not affect data access in any way. It makes relational databases model structure independent.

4. Data Consistency & Accuracy
Since multiple level checks and constraints are built-in, data is accurate and consistent.

5. Easy Data Retrieval & Sharing
Data can be easily extracted from one or multiple relations. Data can also be easily
shared among users.

Physical Data Models
Physical data models describe the storage of data in a computer. They represent information such as record structures, record orderings, and access paths. There are not as many physical data models as logical data models. The most common physical data models are the unifying model and frame memory. In a relational model, every relation or table consists of many tuples. Tuples are also called records or rows.

• Database Environment