Reliability-Driven Fault Tolerant Scheduling Heuristics for Distributed Embedded Real-Time Systems
Salim Kalla, Hamoudi Kalla, Chafik Arar · International Journal of Computer Applications · 2011
a new scheduling approach to produce automatically a fault tolerant distributed schedule for critical distributed and real-time embedded systems. The approach that we propose take as input a set of operations (tasks), a target distributed architecture, some distribution constraints, some indications on the execution times of the operations on the processors of the target architecture, some indications on the communication times of the data-dependencies on the media communications of the target architecture, and the reliability of processors. IT produces a fault-tolerant distributed and static scheduling of the operations on the architecture, with an indication whether or not the real-time constraints are satisfied. The scheduling approach that we propose for architectures with multiple processors linked by a set of channels (links), consist of a list scheduling heuristic based on active replication strategy. In order to reduce the probability of fault occurrence, the replication process of each operation is based on a Global System Failure Rate (GSFR) objective function. Finally, simulation results show the performance of our approach.