Quality Assurance: Software Quality Assurance Made Easy pdfdrive com

Chapter 7: Software Reliability Testing
Software reliability testing relates to the ability of the software to function under
environmental conditions within a period through testing. It helps discover errors
in the software functionality and design.
Software reliability follows a formula. It is: Probability = # of failing cases /
total # of cases under consideration To test for software reliability, data
gathering occurs during the different development stages. Because of restrictions
in time and cost, testing may be limited, as well. If there is sufficient
information, statistical studies commence using fixed deadlines or dates to
handle time constraints. Software design ceases and actual implementation
begins. It is important for data gathering to be precise because of time and cost
restrictions.

---

Software Failure Mechanisms
Software failures are due to misinterpretation of software requirements,
oversights, ambiguities, or errors. Other causes of software failures include
incompetence or carelessness of application developers, unexpected or incorrect
software usage, inadequate testing, and other unexpected problems.
Usually, a defect in the software design can cause failures. Failures can occur
unexpectedly. In some cases, restarting the software can help fix the problems.
However, the reliability of the software does not depend on its life cycle and
time. Environmental factors can affect inputs but they cannot affect reliability of
the software.
It is impossible to predict software reliability. Identical software components
cannot improve software reliability. However, standardized logic structures in
software code, if extensively tested and well understood, can improve reliability
and maintainability.

---