Fault-tolerance through scheduling of aperiodic tasks in hard real-time multiprocessor systems

Abstract

Real-time systems are being increasingly used in several applications which are time critical in nature. Fault-tolerance is an important requirement of such systems, due to the catastrophic consequences of not tolerating faults. In this paper, we study a scheme that provides fault-tolerance through scheduling in real-time multiprocessor systems. We schedule multiple copies of dynamic, aperiodic, nonpreemptive tasks in the system, and use two techniques that we call deallocation and overloading to achieve high acceptance ratio (percentage of arriving tasks scheduled by the system). This paper compares the performance of our scheme with that of other fault-tolerant scheduling schemes, and determines how much each of deallocation and overloading affects the acceptance ratio of tasks. The paper also provides a technique that can help real-time system designers determine the number of processors required to provide fault-tolerance in dynamic systems. Lastly, a formal model is developed for the analysis of systems with uniform tasks.

Fault Tolerance In Grid Computing: State of the Art and Open Issues

“…In PB-overloading primary of a task is scheduled onto the same or overlapping time slot with the backup of another task on a processor. While in BB-overloading backups of multiple tasks are scheduled onto the same or overlapping time slot on a processor [43]. Since PB-overloading can assign an earlier start time than that of the BB-overloading, thus increasing the schedulability In [44], R. Al-Omari et al concluded that the PB-overloading is able to achieve better performance than BB-overloading, and BB-overloading algorithm is better than no-overloading.…”

Fast Algorithms for Testing Fault-Tolerance of Sequenced Jobs with Deadlines

“…It can be shown that the method from [GMM97] does not work for arbitrary release times. Our general approach is similar to those in the previous section.…”

“…Recall that, as in [GMM97], we assume that ∆ ≥ 2p max . As in the previous section, the idea is to show that only some special fault sequences, the cruel fault sequence for GAP ∆ need to be considered.…”

“…The algorithm from [GMM97] verifies fault-tolerance if all jobs are ready at the same time. It can be shown that the method from [GMM97] does not work for arbitrary release times.…”

Algorithms for testing fault-tolerance of sequenced jobs

“…Recall that, as in (Ghosh et al 1997), we assume that Δ ≥ 2p max . We give a linear-time algorithm for testing fault-tolerance in this model in the case of exposed faults.…”