A Data Related Behaviors Automatic Detection Method for Parallel Software Testing
Tao Sun, Wenjie Zhong · 2018
Concurrent software testing is particularly challenging because such systems have a miraculous number of possible executions. Colored Petri Net (CPN) is suitable for parallel description and path analysis and selection, so concurrent system testing can be conducted by means of CPN model. However, CPN has state explosion problem, which makes test coverage difficult to achieve. In this paper, the purpose of a test is described by specifying the group of behaviors to be tested. It is believed that the test coverage can be achieved by covering all paths formed by tested behaviors and data dependent behaviors, and the data irrelevant behaviors will not affect the occurrence of the behaviors to be tested. Therefore, this paper proposes a method for automatic detection of data related behaviors related to the tested behavior, which can automatically detect related behaviors according to the transfer relationship of data in the model. The ultimate purpose of this method is to prepare for the serialization of unrelated behaviors, thereby reducing the state space and improving the test efficiency. In order to realize the method of automatically detecting the data related behaviors, the related data of the behavior to be tested is obtained first. Then, by using the recursive method, the behavior that affects the related data is determined as the data related behavior of the tested behavior. At last, a practical example is presented to verify the effectiveness of the method.