Task Synthesis for Control Applications on Multicore Platforms

Multicore processors promise to improve the performance of systems, by integrating more and more cores onto a single chip. Existing software systems, such as control software from the automation domain, need adjustments to be adapted on multicores. To exploit the concurrency offered by multicore processors, appropriate algorithms have to be used to divide the control application software into tasks, and tailored task partitioning and scheduling approaches are required to increase the overall performance. In this paper we present a model-driven approach for automatic synthesis and deployment of control applications on multicore processors. The approach is centered around a system model, which describes the control applications, the multicore platform, as well as the mapping between the two. We apply the approach on a number of control applications out of which we synthesize tasks and present their run-time behavior in a real-time operating system.