Enhancing Testing of MASs with a Simulator of JADE

In both academia and industry, multiagent systems (MASs) are increasingly employed to model and implement complex distributed applications, due to their natural alignment with intelligent, autonomous, and interactive entities. However, the development of large-scale MAS-based solutions introduces challenges related to maintainability, testability, and performance. Real-world deployments involving numerous agents and distributed devices require thorough testing under conditions that closely approximate the final operational environment, but such testing can be costly in terms of both computational and physical resources. While MAS simulation tools could alleviate this burden, their employment can be tricky and time consuming. In this paper, we introduce SimJade, a simulation framework based on the PeerSim peer-to-peer simulator, designed to execute on a single machine physically distributed JADE systems, without requiring multiple JADE installations or the presence of physical devices. SimJade enables the simulation of complex multi-platform agent-based systems without requiring any modification to the original JADE source code, thereby allowing the testing of a realistic emulation of field deployments. JADE was selected as the reference platform due to its widespread adoption, particularly in industrial contexts where reliable and scalable testing is critical. We demonstrate the effectiveness of SimJade through a series of evaluations involving a real-world MAS intended for deployment across thousands of end-user devices, reporting promising results in terms of scalability and fidelity.