QoS-aware fault-tolerant rate-monotonic first-fit scheduling in real-time systems

Abstract

In real-time systems, fault-tolerant techniques play an important role in guaranteeing the service level agreement (SLA). Primary-backup scheduling is widely used in providing fault tolerance while reducing hardware redundancy. Most existing fault tolerance scheduling algorithms only take fault tolerance performance into account while ignore other metrics, such as quality of service (QoS), load balancing and energy efficiency. Motivated by this fact, this paper proposes a QoS-aware fault-tolerant rate-monotonic first-fit scheduling algorithm, QFTRMFF, consisting of two steps. The first step is to assign tasks to processors using existing real-time fault-tolerant scheduling algorithm. In the second step, a new strategy is put forward to improve the QoS levels of tasks. Experimental results show that compared with other baseline algorithms, the proposed algorithm can substantially improve both quality of security and resource utilization under different task load.