Page 247 - Information_Practice_Fliipbook_Class11
P. 247
Request
Information
Database
DBMS Data
Request Software
Information
Fig 9.2: User interface with DBMS software
9.1.2 Need for DBMS
In the olden days, the data was managed manually in registers or as papers in files. Related files were physically placed
in folders, and folders were organised in cupboards or wardrobes. If we wish to manage all the information about a
company manually, we would need separate file folders to maintain data about employees, departments, projects, etc.
The data about an employee would be typically fragmented in several files. Further, there may be multiple copies of
the same data in different files. For example, an employee's leave record may be maintained by the local office and the
head office. Similarly, the human resources (HR) department and the accounts department may maintain their copies
of the salary payable to the employees. Multiple copies of the same data will lead to the possibility of inconsistency
in data as it may not be possible to apply updates to all copies of the data simultaneously. Thus, as data increases,
ensuring the consistency of data becomes cumbersome.
Over time, manual file systems were replaced by computerised file systems. Searching or modifying data became much
faster in these systems. Based on the computerised file systems, various information systems were developed to meet
the requirements of businesses. However, these systems were also plagued with the problem of data redundancy
because they still contained multiple copies of the same data. Data redundancy leads to the following problems:
1. Wastage of Storage Space: The Same data appears at various places.
2. Increased Data Processing Effort: An update operation involving a data item that appears in several files requires
updating all those files. For example, if an employee's salary is updated, all the files containing information about
the employee's salary will have to be updated.
3. Inconsistent Data: While updating data, it may happen that one file correctly reflects updated data whereas the
other file still contains old data or incorrectly updated data. For example, suppose an employee's salary needs to
be increased by 3500. Using a conventional file system, a programmer may carry out the update correctly in the
file relating to the accounts department. But the salary mentioned in a file in the HR department may not have
been updated. There could be another situation that the salary in the HR department's file might be increased
by only 3000 because an incorrect version of the data file was used for the update. This kind of anomaly is quite
likely to happen if different individuals are responsible for updating the same data at multiple places.
Apart from data redundancy, computerised file systems have several other disadvantages as mentioned below:
1. Structural Data Dependence: A change in the structure of a file may require changes in the applications programs
that use this file. For example, the inclusion of a new attribute email-id in a file containing employee data will lead to
changes in the programs that use the employee file. Similarly, a change in data characteristics, such as an increase
in the length of the address line, will affect other programs that use the employee file. Thus, application programs
become dependent on the file structure.
2. Lack of Controls: Typically, organisations enforce a hierarchical system for defining access rights of different categories
of users. For example, while the head of the EDP department may access all or most of the information available
in an organisation, the head of the HR department, staff of the HR department, head of the accounts department,
and accounts clerk will have restricted access to data. It is hard to enforce rules for authorising legitimate access to
data in applications developed using conventional file systems.
Database Concepts and the Structured Query Language 233

