IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.4, April 2009
230
Manuscript received April 5, 2009
Manuscript revised April 20, 2009
Efficient Testing of Database Applications
Mirza Mahmood Baig† and Ansar Ahmad Khan††
NED University of Engineering & Technology, Karachi Pakistan
Summary
This concept is quite recent provides a technique for more
efficient testing of database applications. In the testing technique
the original database remains intact throughout the experiment.
When tester tests the test suites insert, delete, and modify updates
are physically stored in another file or database called
“Differential Table. The joint operation on Original database and
differential table by sql query language together forms another
database called, “Hypothetical database state”. In this paper, we
have designed a computer aided formula for test suites which
make “decision”, that the test requirement be executed
hypothetically or traditionally. We have also designed pseudo
code for hypothetical inserting and deleting tuple in the database
without changing the originality of the database.
Key words:
Software Testing, Hypothetical database, Differential Table,
Parent Table, traditional database
1. Introduction
The concept of hypothetical relation and differential file
was first explored by Michael Stonebraker as described in
[1, 2, and 3]. The implementation of hypothetical relation
supports “what-if database” which is implemented using
differential file. As per description in [4, 5, 6], a new
concept in the field of hypothetical database namely
“Independent Update Views” popularly known as IUV for
database has been presented. As describe in [7] introduces
terminology and basic concept to extend IUVs for version
management. It forms a version hierarchy and called it
version tree like B tree and represents each node (or child)
is an IUV change version of its parent.
The advantage of hypothetical database testing is that
it is cost effec