Database Models
There are three types of database models which are common in
the industry. They are the
Hierarchical Data Model, the Network Data Model and the Relational Data Model.
These models have their own particular nature, and, therefore, they are
applicable in the industry where a similar situation exists. There is fourth
model known as the Object Data Based Model (ODBM) in which data is structured as
objects at a much higher level as compared to the other models providing
flexibility in designing the application.
The HDBM is applicable when the data in an organization can
be put down in the hierarchical or in terms of the levels, one after another.
The data model is equivalent to a tree.
A tree has roots, branches and leaves; their equivalents in the HDBM begins
records, nodes and fields.
The typical characteristics of the HDBM are:
HDBM starts with a root and has several roots.
A root will have several branches.
Each branch is connected to one and only one root.
A branch has several leaves and a set of leaves are connected
to one branch.
The NDBM interconnects the entities of an organization into a
network. The data model is shown by
an arrangement of the blocks. The
lock represents an entity or a record.
The collection of the blocks is called as the Area of database. The NDBM uses the blocks, the area and
the arrows to represent the database of the organization.
In the RDBM, the concept of two dimensional tables is used to
show the relation. In our example of
the product database, the table showing the component name and component number
is a representation of the data in a table form, RDBM model user's theories of
relational algebra in representing the data in various tables.
The main difference between three models of the database lies
in representation of the entity relationships and its structure, and the user
has to understand the structure, while designing the application based on the
database models.
Of the three, the Relational Model has founded wider
acceptance for the following reasons:
Simplicity: The entity
relationships are identified in simple tabular form, understandable to the users
of the data. It is not related to
any structure of entities like the hierarchy or the network.
Non-procedural requests: Since there is no structure
dependence, the information can be obtained from any point in the database. No
procedural language is to be used.
Data independence: Since the model is based on the relations and
not on the structures, there is a high degree of data independence. This,
however, calls for ensuring that the relationship must be complete and accurate.
Theoretical foundation: The relational data model is based on
the set theory of relations and is designed on the principle of normalization.
For the other models, such mathematical foundation does not exist. The disadvantages of the relational model
are that it demands a large amount of storage, access and speed capabilities to
handle the model.
