Published: October 16, 2012
Author(s)
L. Yu, Yu Lei, Raghu Kacker, Richard Kuhn, J. Lawrence
Conference
Name: 2012 17th IEEE International Conference on Engineering of Complex Computer Systems (ICECCS 2012)
Dates: 07/18/2012 - 07/20/2012
Location: Paris, France
Citation: Proceedings of the 2012 17th IEEE International Conference on Engineering of Complex Computer Systems (ICECCS 2012), pp. 220-229
Combinatorial testing has been shown to be a very effective testing strategy. Most work on combinatorial testing focuses on t-way test data generation, where each test is an unordered set of parameter values. In this paper, we study the problem of t-way test sequence generation, where each test is an ordered sequence of events. Using a general labeled transition system as the system model, we formally define the notion of t-way sequence coverage, and introduce an efficient algorithm to compute all valid t-way target sequences, i.e., sequences of t events that must be covered by at least one test sequence. We then report several algorithms to generate a set of test sequences that achieves the proposed t-way sequence coverage. These algorithms are developed as the result of a systematic exploration of the possible approaches to t-way test sequence generation, and are compared both analytically and experimentally. The results show that while these algorithms have their own advantages and disadvantages, one of them is more scalable than others while exhibiting very good performance.
Combinatorial testing has been shown to be a very effective testing strategy. Most work on combinatorial testing focuses on t-way test data generation, where each test is an unordered set of parameter values. In this paper, we study the problem of t-way test sequence generation, where each test is...
See full abstract
Combinatorial testing has been shown to be a very effective testing strategy. Most work on combinatorial testing focuses on t-way test data generation, where each test is an unordered set of parameter values. In this paper, we study the problem of t-way test sequence generation, where each test is an ordered sequence of events. Using a general labeled transition system as the system model, we formally define the notion of t-way sequence coverage, and introduce an efficient algorithm to compute all valid t-way target sequences, i.e., sequences of t events that must be covered by at least one test sequence. We then report several algorithms to generate a set of test sequences that achieves the proposed t-way sequence coverage. These algorithms are developed as the result of a systematic exploration of the possible approaches to t-way test sequence generation, and are compared both analytically and experimentally. The results show that while these algorithms have their own advantages and disadvantages, one of them is more scalable than others while exhibiting very good performance.
Hide full abstract
Keywords
combinatorial testing; t-way sequence coverage; test sequence generation
Control Families
None selected
) or https:// don't prove our archive is authentic, only that you securely accessed it. Note that we are working to fix that :)
