Distributed T-Way Test Generation Strategies Using Tuple Space Approach
When generating a t-way (where t indicates the interaction strength) test suite for large and complex software systems, the number of interaction between software components to be covered for higher order t-way is likely to be huge and potentially leads towards a combinatorial explosion problem. Apa...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/44034/ http://eprints.usm.my/44034/1/Zainal%20Hisham%20Bin%20Che%20Soh24.pdf |
| Summary: | When generating a t-way (where t indicates the interaction strength) test suite for large and complex software systems, the number of interaction between software components to be covered for higher order t-way is likely to be huge and potentially leads towards a combinatorial explosion problem. Apart from being an NP complete problem, the computational complexity for t-way test suite generation also grows rapidly as the value of t increases. The resultant test case number in test suite also increases exponentially as the value of interaction strength, t is increases. In this manner, t-way test suite generation with large input parameter and high interaction strength require significantly high computational power and memory spaces.
A myriad of useful t-way test suite generation strategies have been implemented recently using the sequential algorithm on standalone machines. Although helpful, the computational power and memory space of a standalone machine is arguably insufficient especially when dealing with large input parameters and high interaction strength. Furthermore, most of available strategies on t-way test suite generation cannot extend the computing work from standalone machines into a multiple machine environment. |
|---|