Fault-Tolerant Scheduling Algorithms Based On Energy Utilization In Heterogeneous Systems

As heterogeneous systems have been deployed widely in various fields,the reliability become the major concern.Thereby,fault tolerance receives a great deal of attention in both industry and academia,especially for safety critical systems.Such systems require that tasks need to be carried out correctly in a given deadline even when an error occurs.Therefore,it is imperative to support fault-tolerance capability for systems.Scheduling is an efficient approach to achieving fault tolerance by allocating multiple copies of tasks on processors.While achieving fault tolerance,consideration should also be given to the use of energy.How to use energy is also the key point of research.Such problems are studied in depth in this thesis.The main work is as follows.First,research on energy efficiency.A fault-tolerant scheduling algorithm considering energy utilization is proposed,DRB-FTSA-E.A multi-core heterogeneous system is used to model tasks using DAG task graph.The reliability of the computing system is studied based on the general exponential distribution assumption of reliability.An energy consumption model which has been applied in other energy management research is used to calculate the energy consumption of the system.The algorithm adopts the active replication strategy,completes a series of tasks under given reliability and time constraints,and achieves high energy utilization.The algorithm finds out a variety of schemes that satisfy the reliability and time constraints,and chooses a scheduling scheme with the highest energy utilization rate as the final scheme.The simulation results show that the proposed algorithm can satisfy the reliability and time constraints,and achieve the highest energy utilization rate.Second,to further optimize the algorithm DRB-FTSA-E.DVFS technology aims to reduce energy consumption.When the clock frequency is too low,the system takes too long to execute the task,which leads to excessive energy consumption or system performance deterioration.Therefore,it is necessary to select the appropriate voltage frequency according to the specific performance of the task.Under given reliability and time constraints,the algorithm SAUE uses DVFS technology to quantitatively calculate the number of fault-tolerant tasks within the time deadline,further reducing energy consumption and maximizing energy utilization.