Day 2: Mini-tutorial: Challenges to the design and optimization of cyber-physical systems

Summary form only given. We are witnessing an exponential increase of cyber-physical systems where the computational components interact with the physical world in a tightly manner. More and more of these systems are nowadays used for safety-critical applications, such as automotive electronics and medical equipment. These safety-critical applications impose stringent requirements on reliability, efficiency, low-power and testability of the underlying VLSI hardware implementation. With silicon technology scaling, however, VLSI circuits is built with smaller transistors, perform at higher clock frequencies, run at lower voltage levels, and operate very often at higher temperature. All these have major negative impact on reliability, performance, power-efficiency and testability. We are therefore facing the challenges of how to address all these technical problems and their interplay with the stringent real-time requirements imposed by many safety-critical applications. This talk will discuss the design of such cyber-physical systems by considering both fault-tolerance and real-time requirements at the same time. It will describe several key challenges and some emerging solutions to the design and optimization of such systems. In particular, it will present time-redundancy based fault-tolerance techniques to address transient faults which have become more and more common in nano-scale technology. It will also describe several design tradeoffs including hardware/software co-design solutions for the optimization of cyber-physical systems.