Weighted parametric systems: Modelling and architecture
share
Modern systems' design usually lies in multiple components which are connected via their interfaces. Synchronous applications, like IoT, require parametric systems, i.e., systems that aim to operate independently of the number of their components. A key feature in modelling of parametric systems is the architecture modelling that defines the underlying topology and provides coordination rules. On the other hand, to achieve optimized requirements and performance there is need for describing parametric architectures and parametric behavior in a quantitative setting. In this paper we study the quantitative modelling of architectures of component-based parametric systems. We introduce a weighted extended first-order interaction logic as a modelling language for weighted parametric architectures. We also introduce weighted parametric models defined by composed transition systems whose interactions are expressed by formulas of our weighted logic. Equivalence of weighted extended first-order interaction logic formulas is decidable in the class of computable fields. We show that our weighted logic can efficiently describe several software architectures with quantitative characteristics.
READ FULL TEXT