Demonstration of model-driven performance prediction of distributed real-time embedded systems of systems

Abstract

Investigating the performance of system of systems raises significant challenges due to their distributed and interconnected nature. To determine the suitability of different architectural alternatives, the proposed performance prediction process captures and analyses system metrics. Metric visualisation provides feedback to system experts who can pose performance questions and decide on the optimal architectural design. To support this process, this paper presents a model driven engineering and system execution modelling tool set. This provides early insight into architectural characteristics and assists in the design and development of Distributed Real-time and Embedded systems of systems.