A novel real-time fault-tolerant scheduling algorithm based on distributed control systems
Hanming Chen, Wei Min Wang, Wei Luo, Jun Xiang · 2011
In the context of real-time fault-tolerant scheduling in Distributed Control systems, Primary-backup scheme plays an important role. A backup copy is always preferred to be executed as passive backup-copy whenever possible because it can take the advantages of backup copy de-allocation technique and overloading technique to improve schedulabilty. In this paper, we propose a novel Fault-tolerant Rate-Monotonic Best-fit algorithm (NRMBF) based on Distributed Control systems to enhance the schedulability. Unlike existing scheduling algorithms (FTRMBF or FTRMFF) that will allocate a new processor for backup copy when no existing processors can accommodate the current backup copy to be assigned. NRMBF uses a back tracing strategy that reassign the primary copy when the aforementioned situation occurs. As a result, NRMBF can use fewer processors to schedule a set of tasks without losing real-time and fault-tolerant capability of the system. Simulation results reveal that NRMBF significantly improves the schedulabilty over existing, comparable algorithms in literature.