Using Multicore Architectures in Cyber-Physical Systems

The demand for higher performance computing platforms has dramatically increased during the last decade due to the continuous feature enhancement process. For instance, in automotive systems new safety features like `night view assist’ and `automatic emergency breaking’ require the fusion of sensor data, video processing and real-time warnings when an obstacle is detected on the road; in the avionics domain new applications such as the helmet-mounted display systems require intensive video processing capabilities. Commercial-Off-The-Shelf (COTS) components are increasingly used in an effort to raise performance and lower production costs. Fast multicore CPUs and high-performance DMA peripherals are needed for servicing these new demanding applications. However, they are difficult to use due to problems with timing predictability and security. A multicore architecture is substantially different from a single core implementation in that concurrently executing tasks (on the parallel cores) share critical physical resources such as caches, peripherals, on-chip interconnect networks, etc. This extensive sharing of physical resources on critical paths can jeopardize the timing predictability and at times, security, of safety-critical software applications even when system resources are under the control of a real-time operating system. The stark reality is that without addressing these issues, high assurance product developments will be unable to take full advantage of emerging multicore CPUs.