Improving stress search based testing using a hybrid metaheuristic approach
Francisco Nauber Bernardo Gois, Pedro Porfírio Muniz Farias, André L. V. Coelho, Thiago Monteiro Barbosa · 2016
Some software systems must respond to thousands or millions of concurrent requests. These systems must be properly tested to ensure that they can function correctly under the expected load. A common use of stress testing is to find test scenarios that produce execution times that violate the timing constraints specified. In this context, search-based testing is seen as a promising approach for verifying timing constraints. The main purpose of this paper is determine if hybrid algorithms are superior to single metaheuristics in search-based stress testing. The proposed hybrid metaheuristic approach uses genetic algorithms, simulated annealing, and tabu search algorithms in a stress testing model. The secondary objective of this paper is to improve stress testing automation. A tool named IAdapter, a JMeter plugin used for performing search-based stress tests, was developed. Two experiments were conducted to validate the proposed approach.