Algorithms Of Fault-Tolerant In Distributed Real-Time Systems

In order to safeguard the reliability of distributed real-time systems, the primary/backup duplication technique is adopted to improve the system’s fault-tolerance. Base on this, combining the task allocation algorithm named first-fit, a fault-tolerant scheduling algorithm for heterogeneous distributed real-time systems is presented. Integrating three characteristic parameters (i.e. deadline、value、slack) of real-time tasks to calculate the priority of the task, and the real-time tasks are sorted according to priority. Based on the analysis of distributed real-time systems, system fault tolerance model is given. On this basis, the task fault-tolerant scheduling conditions are given. Under the premise of the task fault-tolerant scheduling, the deadline of the primary copy is optimized in order to balance processor load and improve processor utilization. Finally, through the simulation experiments, this algorithm is compared with the first-fit algorithm. Simulation results show that the proposed algorithm enhance the task receiving rate and fault-tolerant performance, you can get more than the first-fit algorithm higher processor utilization.