Sensitivity-driven co-synthesis of distributed embedded systems

This paper describes a new, sensitivity-driven algo-rithm for the co-synthesis of real-time distributed em-bedded systems. Many embedded computing systems are distributed systems: communicating periodic pro-cesses executing on several CPUs/ASICs connected by communication links. We use performance estimates to compute a local sensitivity of the design to process allocation. We propose a priority prediction method to schedule processes. Based on these techniques, we de-velop a gradient-search algorithm which co-synthesizes a heterogeneous distributed systems of arbitrary topol-ogy and the associated application software architec-ture. Experimental results show that our algorithm can nd good implementation architectures in small amounts of CPU time.