Database oriented approach of Organising data provides a set of programs to facilitate users to organise, create, delete, update and manipulate data in a database. All these programs together form a Database management system ( DBMS). Features found commonly in database management system are described below :
Database models:
We saw that database integrates together multiple related files. A database model defines the manner in which a database links it's various data elements tougher. Four database models used commonly are, hierarchical, network, relational, and object oriented. Also known as database structure techniques, they are described below.
1. Hierarchical Database:
A hierarchical database links it's data elements as an inverted tree structure ( root at the top with branches formed below). Below the single root database element are subordinate elements, each of which, in turn has its own subordinate elements and so on, the can grow to multiple levels. Database elements of a hierarchical database have parent child relationship as in a family tree. A parent data element is one that has one or more subordinate data elements. Data elements that are below a parent data element are it's children data elements. There may be many children data elements under each parent element, but there can be only one parent element for any child element. Note that the branches of a tree structure are not connected.
Several database applications use hierarchical database because they can neatly organise their data elements as a hierarchical tree structure. However, the main limitations of hierarchical database is that it does not support flexible data access. In a hierarchical database, applications can access it's data elements only by following paths formed by branches of the tree structure. Hence, a database designer designing a hierarchical database must determine all types of relationships among the data elements and map them properly in a hierarchical tree structure while designing the database.
2. Network Database:
A network Database is an extension of hierarchical database. A network Database also organises it's data elements I such a manner that they have parent child relationship among them. Like I'm hierarchical database, a database designer designing a network Database must determine all types of relationships among data elements while designing the database. In a network Database, however, a child database element can have more than one parent element or no parent element at all. Moreover in this type of Database, database management system permits extraction of needed information by beginning form any data element in database structure. Instead of starting form root data element.
3. Relational Database:
A relational database organises it's data elements as multiple tables with rows and columns. It stores each table as a separate file. Each table column represents a data field, and each raw represents a data record (also known a as tuple). A common field links data in one table or file to another table.
Relational database model provides greater flexibility of data organisation and future enhancement in database as compares to hierarchical and network Database models. Many applications can organise their data elements in a relational database in a manner that is identical to real life relationships between data elements. Hence, users often find relational database model to be closer to their intuitive model of real life situations, making making this model easier to use and understand. Moreover for adding new data to an existing relational database, there is no need to redesign the database afresh. Rather, user can add to the database new tables containing new data and can then link this tables to existing tables by common key fields. Users can also recognise data elements, when necessary, to create new table by selecting certain rows or specific columns from other tables or by joining columns and rows form two separate tables.
4. Object Oriented Database:
Hierarchical, relational, and network data models are suitable for conventional business data processing applications, such as inventory control, payroll, accounts, employee information systems, etc. An application designers have found them unsuitable for a variety of other types of applications because of several serious shortcomings in them. This applications include computer-aided design( CAD), computer-aided engineering (CAE), computer-aided manufacturing (CAM), computer-aided software engineering (CASE), expert systems and multimedia systems.
Object Oriented Database model overcomes this shortcomings of conventional database models. An object oriented database is a collection of objects whose behaviour, state, and relationships are in accordance with object oriented concepts ( such as abject, class, class hierarchy, etc.). An object oriented database management system allows definition and manipulation of an abject oriented database.
Comments
Post a Comment