Database Design Practices

In database discussions, it is common to hear the term normalization or database design come up. However,for the most parts these discussions revolve around splitting up the data that is stored between different tables to improve performance and to eliminate duplication.  This is extremely important and should be discussed in any discussion on tables but for the most part the discussions do not involve splitting up data between databases.  This is either assumed or just not put into practice as often as it should be.  Either way, it is important to consider when designing an application that will need to access data.

Information Commonality Data could be defined as the actual values that are stored in the database itself and are static until they are changed by some process (either manually or automated). Data by itself is useless if not processed in such a way as to give it meaning.  Information, on the other hand, is data processed in such a way as to give it meaning and is constantly changing based on the data stored.  In addition, the majority of data collected can be processed in multiple ways to provide multiple meaningful outputs.  In databases, many developers look at the data that is contained versus the information that the data is designed to provide.  Databases should be developed in such a way that each database produces a common set of information.  Tables that would hold data associated with human resource information should not be contained in the same database as tables that would contain parts for aircrafts in an airline database system.  If there is a need for the data in the inventory database to interact with data in the human resource database to provide the proper information, the MySQL server can pull data from the tables holding the data (even though they are in two different databases) to provide such information.  

Exclusive Data

In such situations where there are separate entities that utilize the same data structures, separating out the data may be the safest and most reliable way to ensure data integrity.  An example would be an accounting firm that is responsible for the payroll of two or three companies.  The application that would be created to support such as system could utilize the same data structures but have separate databases containing the actual data for each company.  This would eliminate any chance that data from one company could interact

with data from another company which would create an insecure and potentially precarious situation.  This would require time and energy put into ensuring that the database and its underlying components were solid enough to avoid the need for structural changes after implentation. Copying database structures from one database to another is an easy process in MySQL; however, any future changes would have to be done in the individual databases separately.