THE DATABASE APPROACH - Database Design and Development: An Essential Guide for IT Professionals - page 16

file-oriented systems. Data and processing logic had to be separated so as to
improve programmer productivity. The new approach of using databases instead of
conventional flat files addressed the challenges for meeting the increased demand
for information. The database approach overcame the limitations of the earlier data
systems and produced enormous benefits. Let us review the specific benefits and
understand in what way the database approach is superior to the earlier data
systems.
Minimal Data Redundancy Unlike file-oriented data systems where data
are duplicated among various applications, database systems integrate all the
data into one logical structure. Duplication of data is minimized. Wastage of
storage space is eliminated. Going back to the bank example, with a database,
customer data is not duplicated in the checking account, savings account, and loan
account applications. Customer data is entered and maintained in only one place in
the database.
Sometimes, in a database, a few data elements may have to be duplicated. Let us
say that product data consist of product number, description, price, and the corre-
sponding product line number. All the fields relating to product line data are kept
separately. Whenever the details of products and product lines are needed in appli-
cations, both data structures are retrieved from the database. Suppose a heavily used
product forecast application needs all the details of the product from product data
and just the product line description from the product line data. In that case, it will
be efficient for the product data to duplicate the product line description from the
product line data. Thus, in some instances, data duplication is permitted in a data-
base for the purpose of access efficiency and performance improvement. However,
such data duplications are kept to a minimum.
Data Integrity Data integrity in a database means reduction of data inconsistency.
Because of the elimination or control of data redundancy, a database is less prone
to errors creeping in through data duplication. Field sizes and field formats are the
same for all applications. Each application uses the same data from one place in the
database. In a bank, names and addresses will be the same for checking account,
savings account, and loan applications.
Data Integration In a database, data objects are organized into single logical data
structures. For example, in file-oriented data systems, data about employees are scat-
tered among the various applications. The payroll application contains employee
name and address, social security number, salary rate, deductions, and so on. The
pension plan application contains pension data about each employee, whereas the
human resources application contains employee qualifications, skills, training, and
education. However, all data about each employee are integrated and stored
together in a database.
So, in a database, data about each business object are integrated and stored sep-
arately as customer, order, product, invoice, manufacturer, sale, and so on. Data inte-
gration enables users to understand the data and the relationships among data
structures easily. Programmers needing data about a business object can go to one
place to get the details. For example, data about orders are consolidated in one place
as order data.