Information Systems Reengineering, Integration and Normalization - Information Systems Reengineering, Integration and Normalization - page 15

1.3.3 Integration of Multiple Databases

There has been a proliferation of databases in most organizations. These databases
are created and managed by the various units of the organization for their own
localized applications. Thus, the global view of all the data that is being stored
and managed by the organization is missing. Schema integration is a technique to
present such a global view of an organization's databases. There has been a lot of
work done on schema integration. ￖzsu and Valduriez ( 1991 ) presented surveys of
work in this area. All these techniques concentrate on integrating database schemas
without taking into consideration new database applications. We need a practical
approach to schema integration to support new database applications by compar-
ing the existing databases against data requirements of the new applications. If the
existing databases are inadequate to support new applications, they must then be
evolved to support them.
Since the relational databases emerged, they have been widely used in commer-
cial organizations. However, in an organization, different departments or sections
would have probably developed their own relational database systems according
to their own requirements at various times. Thus, large quantities of data are frag-
mented across a variety of databases. Data could then be redundant and inconsis-
tent. A global view on all data is not possible. This will affect the effectiveness
of decision-making in an organization, as these disparate data do not adequately
support the information needs of an organization operating in a dynamic business
environment. It is vital that a data resource should provide current data for devel-
opment of up-to-date information to support just-in-time decision making in an
organization. There is a great need to create a global view on all existing disparate
data by integrating them in a global database so as to support dynamic and complex
business activities.
Data integration is to implement a global database by integrating various source
databases into a global target database. To accomplish the task of data integration,
the first step is schema integration. This process involves many steps including
solving conflicts between source databases, capturing the semantics of entity, weak
entity, cardinality, isa, generalization, categorization and aggregation of the rela-
tions, and merging to a new integrated schema for each pair of the existing rela-
tional schemas in the source databases.
The next process is data integration. Its objective is to merge data from source
databases to the new global database without any loss of information. It must trans-
form the data structure from the sources to the target-integrated global database
while preserving its semantics. It also uses the data structure of the integrated sche-
ma derived from schema integration.
The integrated global database can be verified by confirming the recaptured seman-
tics from examining its data occurrence. If the recovered semantics matches the seman-
tics of the integrated schema, then the original semantics are preserved in the integrated