A Reliability Model for the Task Scheduling in Distributed Systems based on Fuzzy Theory

Distributed Systems is a mean to run multiple transactions simultaneously. In distributed systems, while dealing with real-time assignments, tasks have to be scheduled correctly and timely. All practical real-time scheduling algorithms in distributed systems present a trade-off between performance and computational complexity. This is mainly due to the complex nature of Mathematical models. In comparison, fuzzy is a simplified approach that optimizes the complete system that too with less time complexity. Further Fuzzy approach produces results that are closer to real world problems. The most innovative part of the research is that reliability metrics have been taken as the major parameter for decision for scheduling. The priority is computed based on the values of Failure rate, CPU time and Reliability. The problem has been simulated on MATLAB 7.0.4 Mamdani Fuzzy Inference Engine to evaluate the performance of the proposed methodology. Experimental results have shown that the proposed fuzzy scheduler creates feasible schedules for homogeneous and heterogeneous tasks.