Wednesday, October 30, 2013

Understanding Client Server and database models

Understanding Client-Server: Client / Server can be interpreted as the ability of the computer to request data request services to other computers. Each instance of the service requesting computer is called a client, whereas each instance that provides the service is called a server. Data requested by the client can be retrieved from a database on the server side that is often called a database server, such as MySQL, PostgreSQL, Oracle, or SQL Server .

The advantages of this client server are more suitable for small businesses.

The database server contains the database engine, including tables, stored procedures, and triggers (which also contain business rules). In the client / server system, most business logic is usually applied in the database. The database server handles:

§ Data management

§ Security

§ Query, trigger, stored procedure

§ Error handling

The client / server architecture is a step forward as it reduces the processing load from the central computer to the client computer. This means more and more users are incremented on client / server applications, server file performance will not decrease quickly. With client / server users dair various locations can access the same data with little load on a single machine.Database Models
Database Management System (DBMS) or database management system is divided into five models. Older models were introduced in the 1960s, which are hierarchical and networked. The newer models are relational, object-oriented, and multidimensional.

Hierarchical Database

In the Hierarchical database, fields or records are organized in related groups, resembling a tree diagram, with the child record (lower level) below the parent record (higher level). The hierarchical database is the oldest and simplest model of the five database models. In this database model accessing or updating data can take place very quickly because the relationships have been determined. However, since the structure must be defined first, then this is quite risky. Furthermore adding new fields to a database record keeps all databases redefined. That's why a new database model is needed to show the problem of data repetition and complex data relationships.

Network Database

The concept of a network database is similar to a hierarchical database but each child record can have more than one parent record. Furthermore each child record can be owned by more than one record parent. Network databases are basically used with mainframes, more flexible than hierarchical databases because there are different relationships between databases. But the structure must still be defined first. Users should be familiar with the database structure. Moreover, the number of inter-record relationships is also limited, and to test a field one must retrieve all records.

Relational Database

The Relational Database works by linking data to different files using a common key or data element.
How the relational database works: The data elements are stored in another table that forms rows and columns. In this database model the data is logically arranged, that is, by content. Each record in the table is identified by a field - the primary key - that contains a unique value. That is why the data in the relational database can appear in a way different from the way it is stored physically on the computer. Users may not know the physical location of a record to retrieve its data.

Object-Oriented Database

This model uses objects as software written in small pieces that can be reused as elements in a database file. Object-oriented database is a multimedia database that can store more data types than a relational database. One object-oriented database model is a hypertext database or web database, which loads text and is linked to other documents. Another model is the hypermedia database, which includes links as well as graphics, sound, and video.
Example: DB2 database, Cloudscape, Oracle9i and so on

Multidimensional Database

Multidimensional Databases (MDAs) model data as fact, dimension, or numerical to analyze large amounts of data, the purpose is to make decisions. The Multidimensional Database uses a cube form to represent the dimensions of data available to a user, a maximum of four dimensions.
Example: InterSystem Cache, ContourCube, and Cognoa PowerPlay

No comments:

Post a Comment