Varying Topology of Component-Based System Architectures Using Metaheuristic Optimization

Abstract

Today's complex systems require software architects to address a large number of quality properties. These quality properties can be conflicting. In practice, software architects manually try to come up with a set of different architectural designs and then try to identify the most suitable one. This is a time-consuming and error-prone process. Also this may lead the architect to sub optimal designs. To tackle this problem, metaheuristic approaches, such as genetic algorithms, for automating architecture design have been proposed. Metaheuristic approaches use degrees of freedom to automatically generate new solutions. In this paper we present how to address topology of the hardware platform as a degree of freedom for system architectures. This aspect of varying architectures has not yet been addressed in existing metaheuristic approaches to architecture design. Our approach is implemented as part of the AQOSA (Automated Quality-driven Optimization of Software Architectures) framework. AQOSA aids architects by automatically synthesizing optimal solutions by using multiobjective evolutionary algorithms and it reports the trade-offs between multiple quality properties as output. In this paper we use an example system to show that the hardware-topology degree of freedom helps evolutionary algorithm to explore a larger design space. It can find new architectural solutions which would not be found otherwise.