Finding Load Inducing Test Scenarios Using Genetic Algorithms and Tree Based Encoding

Abstract

Load test is conducted in order to gain an insight to the characteristics of a system under various amount of load. Since the combination of possible actions a user can follow from start to finish is possibly endless, the possibility of missing a load inducing scenario by using a traditional load testing software is highly probable. In this work, we implement a rule-aided scenario generation algorithm and find the possible scenarios that a high amount of load is generated by using genetic algorithms to drive the search forward.