LWS: A Framework for Log-based Workload Simulation in Session-based SUT

Microservice-based applications and cloud-native systems have been widely applied in large IT enterprises. The operation and management of microservice-based applications and cloud-native systems have become the focus of research. Essential and real workloads are the premise and basis of prominent research topics including performance testing, dynamic resource provisioning and scheduling, and AIOps. Due to the privacy restriction, the complexity and variety of workloads, and the requirements for reasonable intervention, it is difficult to copy or generate real workloads directly. In this paper, we formulate the task of workload simulation and propose a framework for Log-based Workload Simulation (LWS) in session-based application systems. First, LWS collects session logs and transforms them into grouped and well-organized sessions. Then LWS extracts the user behavior abstraction based on a relational model and the intervenable workload intensity by three methods from different perspectives. LWS combines the user behavior abstraction and the workload intensity for simulated workload generation and designs a domain-specific language for better execution. The experimental evaluation is performed on an open-source cloud-native application and a public real-world e-commerce workload. The experimental results show that the simulated workload generated by LWS is effective and intervenable.