Information Systems Reengineering, Integration and Normalization - Information Systems Reengineering, Integration and Normalization - page 7

1.2 The Need

The recent rapid growth in database technology has actually encouraged more in-
stallations of DBMS in different kinds of organizations. In addition to new data-
base system installations, there is considerable interest in converting conventional
file-oriented systems to database systems and upgrading outdated database systems
to a newer database technology. The need to compete, reduce costs, and standard-
ize operational procedures makes conversions to a new technology a necessity for
many organizations. The fact that many large companies still have a large number of
sequential file systems indicates a strong need to convert such systems to a database
system for better management. The introduction of Internet computing makes the
Extensible Markup Language (XML) model a necessity to most companies.
The concept of a relational database was proposed by E. F. Codd in the 1970s.
It is recognized as a more user-friendly model than nonrelational (e.g., hierarchical
or network) models. However, it was not adopted by the industry until the 1980s
because of its poor performance. Thanks to the improvements in their performance,
relational databases have gained wider industry acceptance. These improvements
have created a need to convert data from a nonrelational to a relational structure.
The object-orientated approach to software engineering has recently become
popular, with many manufacturers claiming to have object-oriented systems. Ob-
ject-oriented modeling is a new way of representing static and dynamic data se-
mantics in the form of objects, links, associations, and methods. Traditional record-
based databases (e.g., hierarchical, network, and relational) have been generally
used over the past two decades. Organizations with such record-based databases
could reengineer their databases into object-oriented databases to capture more se-
mantics of the application domain.
Any medium to large organization that has an independent EDP department typi-
cally has a number of databases. Over the past four decades, a number of database
systems have come into the market using these predominant data models: hierarchi-
cal, network, relational, object-oriented, and XML. As a result of this proliferation
of systems, many large organizations have found that they must support various
types of database systems at the same time. However, as the performance of the
relational database systems has improved, they have been accepted by the industry
and consequently created the need to convert a company's nonrelational database
systems to relational.
The hierarchical and network database systems use the concept of currency and
require users to navigate through the database from one point to the next. This
makes them difficult to use for both end-users and programmers because of the
level of skill and experience required to perform this navigation. On the other hand,
a relational database is simpler, as it presents to users relations that resemble files in
a manual cabinet file system.
In the hierarchical and network models, the connections between sets of data are
hard-coded into the data structure and the addition of a new relationship requires
a new access path to be added. In relational databases, access paths are not prees-