Multi-Core Partitioned Scheduling For Fixed-Priority Periodic Real-Time Tasks With Enhanced RBound

This paper presents a new partitioned scheduling approach to schedule fixed-priority periodic real-time tasks on multi-core platforms under Rate Monotonic Scheduling (RMS) policy. We first develop a novel method to scale task periods, as well as execution times, to improve the utilization bound (i.e. RBound [7]). We further apply the task set scaling method and the enhanced utilization bound to multi-core scheduling to improve the system utilization and at the same time guarantees the satisfaction of all timing constraints. We empirically evaluate the proposed techniques and the results show clear evidence of the proposed approach outperforming earlier work existed in the literature.