A new cat swarm optimization based algorithm for reliability-oriented task allocation in distributed systems

Distributed Systems (DS) are usually complex systems composed of various components. Due to increasing complexity and scaling of DSs, reliability becomes a major challenge for the design of such systems. The nodes and links of a DS typically have different hazard rates; therefore, proper task allocation can significantly improve system reliability. On the other hand, optimal task allocation in DSs is an NP-hard problem, thus finding exact solutions are limited to small-scale problems. This paper presents a new swarm intelligence technique based on Cat Swarm Optimization (CSO) algorithm to find near optimal solution. For evaluating the algorithm, CSO is compared with Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The experimental results show that in contrast to PSO and GA, CSO acquires acceptable reliability in reasonable execution time. The confidence interval is set at the 95% confidence level. By assessing the confidence interval, it is observable that CSO has low reliability deviation.