Towards an Automated Approach to Use Expert Systems in Performance Testing

Performance testing in highly distributed environments is very challenging. Specifically, the identification of performance issues and the diagnosis of their root causes are timeconsuming and complex tasks which usually require multiple tools and heavily rely on expertise. To simplify these tasks, hence increasing the productivity and reducing the dependency on human experts, many researchers have been developing tools with built-in expertise for non-expert users. However various limitations exist in these tools, such as managing huge volumes of generated data, that prevent their efficient usage in the performance testing of highly distributed environments. To address these limitations, this paper presents an adaptive framework to automate the usage of expert tools in performance testing. In this paper, we use a tool named Whole-system Analysis of Idle Time to demonstrate how our research work solves this problem. The validation involved two experiments which assessed the accuracy of the proposed adaptive framework and the time savings that it can bring to the analysis of performance issues. The results proved the benefits of the framework by achieving a significant decrease in the time invested in performance analysis.