Relational databases generally follow the work of Codd and Date, and represent data in tables. The tables can be logically joined together dynamically. All "relational" databases use some flavor of SQL for manipulation. There are many other kinds of database systems, which have many of the same characterists and are better or worse for specific applications. Heirarchical databases were popular in the 1980s, they were a logical tree, and much more efficient than RDMBS for many business practices, such as a purchase order, which naturally has a set of line items for parts in the order. Inverted file databases allow generalized queries, but do not use SQL. Object Oriented databases are designed to better map between OO languages such as C++, C#, or Java and databases, since RDBMS make this mapping very complex.

Hi Sunil,

Non-relational databases
 Non-relational databases place information in field categories that we create so that information is available for sorting and disseminating the way we need it. The data in a non-relational database, however, is limited to that program and cannot be extracted and applied to a number of other software programs, or other database files within a school or administrative system. The data can only be "copied and pasted.“ Example: a spread sheet

Relational databases
 In relational databases, fields can be used in a number of ways (and can be of variable length), provided that they are linked in tables. It is developed based on a database model that provides for logical connections among files (known as tables) by including identifying data from one table in another table

Thanks