| Multicore microcontrollers are rapidly penetrating the real-time systems market with a promise of increased processing throughput and lower energy consumption than traditional processors. Consequently, it sparked a new wave of research in soft real-time scheduling algorithms for multicore processors in an attempt to maximize the processing throughput in mass consumer products. The new research has produced many different scheduling algorithms that are good at harnessing the added power of the multicore processors but lack the required determinism to be used in Hard Real-Time systems.;Keywords: Hard Real-Time, Partitioning, Schedulability analysis, Multicore.;We present the Multicore Priority Ceiling Protocol scheduling algorithm for Hard Real-Time systems. Most existing algorithms for multiprocessor Hard Real-Time systems were developed before the multicore era and were intended for distributed systems where an instance of the Real-Time Operating System would be running on each processor. The use of these legacy algorithms therefore requires using multicore chips as loosely coupled architectures or distributed multiprocessor systems through a supervisor losing both design simplicity and effectiveness. To eliminate this unnecessary complexity and to take advantage of the tightly coupled processors of a multicore chip, we developed the Multicore Priority Ceiling Protocol along with its schedulability test for Hard Real-Time systems running on a single instance of a modified Real-Time Operating System. We also developed a tool that uses our schedulability test to solve the task partitioning problem using a Particle Swarm Optimization. |
