Performance analysis of embedded multiprocessor industrial applications: methodology and tools

We present a framework and its supporting simulation tool for modelling and performance analysis of multiprocessor embedded systems. Our framework consists of component-based models for modelling parallel software and multiprocessor hardware, and tools for code generation and performance analysis. The framework component meta-model relies on transaction-level description of hardware and programmer-level description of software where timing properties of the hardware micro-architectures are modelled as annotations in the concurrency meta-model. The framework has the advantage of allowing fast precise and scalable joint analysis of software and hardware performance rather than evaluating each one in isolation. This joint evaluation enables predicting the impact of hardware configuration on software performance and hardware capability to accommodate other services at design time.