A feedback control mechanism for the automation of stress and load testing

Stress and Load Testing are of special interest to many segments of the software industry. However, existing approaches do not provide a solution that is automated, widely applicable, flexible, efficient, accurate, and based on a rigorous theoretical foundation. For example, while some approaches are automated, they cannot be used to stress/load test different types of applications. A novel Stress and Load testing approach is proposed here. It is the first approach to use a proportional integral derivative (PID) Controller to automatically drive the inputs and to achieve a pre-specified level of stress/load for resources of interest. A new adaptive System Identification algorithm is proposed that automatically identifies inputs that impact the resources. A new automatic PID Tuning algorithm is also presented that defines the internal parameters (KP, KI, and K D) that are important to achieve a stable process. The PID Tuning can be done with different goals in mind depending on the Rise Time, Overshoot, Settling Time, and Error at Equilibrium. The approach is rigorously validated with four experiments in addition to multiple simulation runs. The results indicate the new automated approach offers a solution that significantly advances the state-of-the-research in Stress and Load Testing.