Reliability and cost optimization in distributed computing systems

The reliability of the communication network and its processing units and the strategy of task allocation are essential in determining the system reliability of a distributed computing system. Reliability of such systems can be improved by endowing resource redundancy or the use of highly reliable components. In this paper, we develop a relationship between system cost and hardware redundancy levels, assuming cycle-free distributed computing systems. Based on the derived relationship, we propose a hybrid heuristic which combines genetic algorithms and the steepest decent method to seek the optimal task allocation and hardware redundancy policies such that system cost is minimized.