|
|
| Advantages |
| Advantages of a DBMS |
The main advantage of using a DBMS is that the formalism of the model of data underlying the DBMS is imposed upon the data set to yield a logical and structured organization of the data. Given a fuzzy, real-world data set, when a model's formalism is imposed on that data set the result is easier to manage, define and manipulate.
Different models of data lead to different organizations. In general the relational model is the most popular because that model is the most abstract and easiest to apply to data while still being powerful.
Therefore, using a DBMS we have the following advantages
- Clear picture of logical organization of data set
- Centralization for multi-users.
- Data independence
|
| |
| Avoiding Inconsistency |
| Ideally, each piece of data in DBMS should be stored once in a centralized location so inconsistency cannot arise. It is possible to allow duplicate entries but in such cases the DBMS should be aware of this so it can control inconsistency and can guarantee that the user never sees this by propagating updates (insert, delete, modify). Most commercial products do automatic update propagation or controlled redundancy. |
| |
| Eliminating Redundancy |
| Eliminating redundancy is useful for saving storage, but this is a minor point in modern computing. Not all redundancy should necessarily be eliminated even in a logical organization, as sometimes there may be sound business or technical reasons for maintaining multiple copies of the same stored data. However, the DBMS should be aware of where it exists and assume responsibility for propagating updates to all copies. |
| |
| Backup and Recovery |
| In the event of damage (human error, hardware failure, etc) to any portion of the database, it is essential to be able to repair the data concerned with minimum delay and with as little effect as possible on the rest of the system i.e. the availability of data that has not been damaged should ideally not be affected. Therefore the DBA must define and implement an appropriate recovery strategy e.g. periodic dumping etc, by using various utilities and tools. A DBMS product will have in-built into it a mechanics to recover a database into an integral state in the event of system failure. |
| |
| How data is stored |
A data model is defined as a set of guidelines for representing the logical organization of data in the database; a pattern according to which data and relationships can be organized; an underlying mathematical formulation for building logical data organizations.
A data model consists of
- A named logical unit (record type, data item)
- Relationships among logical units
- A data item is the smallest logical unit of data, an instance of which is known as a data item value.
- A record type is a collection of data items, and a record is hence defined as an instance of a record type.
|
| Note : A data model does not specify the data, data implementation or physical organization only the way it can be logically organized |
|
