A Real-Time Fault-Tolerant Scheduling Algorithm of Periodic Tasks in Heterogeneous Distributed Systems

This paper proposes a novel reliability model based on preemptive periodic tasks.Compared with existing reliability models in literature,the proposed reliability model can be one-processor-failed fault-tolerant,which makes it more realistic and precise.Moreover,a real-time fault-tolerant scheduling algorithm based on heterogeneous distributed systems,namely IRDFTAHS,is presented.IRDFTAHS tries to assign tasks copies in a way to improve reliability of system.In addition,IRDFTAHS considers backup copy in both active and passive status,which makes the proposed algorithm more flexible than existing algorithms.Finally,simulation experiments are carried out to compare the algorithm with existing ones in several aspects.The experiments results show that the IRDFTAHS generally performs significantly better than existing algorithms.