Distributed Database is a collection of many logically connected databases and all these databases are located on different locations with the help of any computer network. But the interesting thing is that when we provide view to the user then it is completely transparent and user is blind to see that generated view fetches the data from different databases.
Distributed Database Management System DDBMS
DDBMS is a software that is used to implement and manage the distributed databases.
Promises of DDBSs
Transparency of Distributed Data
Data is distributed on different locations on different servers and only a single view of multiple databases is provided to the user. It means that all this distributed mechanism is transparent or hidden from the user. User can’t guess that data is coming from where.
Data definition of databases occurs at two different levels. One is physical level and another is logical level.
Physical Level – The Physical structure of the data is specified in physical level. Sometimes it is named as physical layer.
Logical Level – The Logical structure of the data is specified in logical level. Sometimes it is named as logical layer.
When we specify physical and logical structure of the data, then it means that we are defining the database schema. Schema definition is also referred as physical data description.
Data independence is also categorized into physical data independence and logical data independence.
Logical data independence – Logical data independence means that any change in the logical schema has no effect on external schema of the database.
Physical data independence – Physical data independence means that any change in the physical schema has no effect on logical schema of the database.
External data independence – It is always independent because there is no level above this level.
Network transparency is a process of hiding the operational details of all the network from the user. Two common transparencies are location transparency and naming transparency.
Same data copies are available on different servers. If user query the database and database is unable to answer the query from one server, then it transfers this query to another distributed database located on different location. All this mechanism is transparent to the user. The user is unaware that from where the data is fetched.
In Distributed database systems, data is fetched by creating different fragments of the database tables. In fragmentation transparency, user is unaware of these fragments.
There are two basic types of fragmentation.
Horizontal fragmentation and vertical fragmentation are used to create the fragments of the tables.
Horizontal fragmentation – In horizontal fragmentation, we use the tuple (rows ) to divide the database into fragments.
Vertical fragmentation – In vertical fragmentation, we use the attributes (column ) to divide the database into fragments.
Table 1: Exam
Table 2: Horizontal fragmentation
Table 3: Vertical fragmentation