Affiliation: Polytechnic Institute of Porto
Title of talk: Real-Time Parallel Models
Abstract: Computers across all domains increasingly rely on multiple cores, with processors starting appearing with dozens to hundreds of cores, and next generations of processors expected to integrate hundreds to thousands of simple cores into a single chip. Real-time and embedded systems are no exception to this trend, and the design of such systems needs to increasingly consider both the set of concurrent tasks and the intra-task parallelism (as it is commonly known), i.e. the possibility of having more than one core executing a single job at any given instant in time. The subject of parallel execution is well-studied in the high-performance domain but most these approaches have inherent limitations for real-time systems as they aim for best-performing, dynamic load-balancing, not time predictability. This talk will present some of the challenges of parallel real-time models, focusing in particular in the integration into the real-time domain of techniques from the high-performance area, such as work-stealing scheduling and parallel programming models.