Discrete binary cat swarm optimization for scheduling workflow applications in cloud systems

There are many challenges evolved in adopting the cloud environment for scheduling workflow applications. The reason behind this is the resource heterogeneity and on-demand services offered by cloud service providers using per-per-use-model. Mapping of the distributed resources to the tasks of an application while satisfying the user's quality of service (QoS) parameters become tedious job because of heterogeneity of resources and dependent nature of the tasks. Many meta-heuristics based approaches are applied by the researchers for scheduling workflow applications to attain near optimal solution. A discrete binary cat swarm optimization (DBCSO) is proposed for scheduling workflow applications in the present work. The purpose of this algorithm is to optimize workflow schedule length also known as makespan. Scientific workflows of different sizes are evaluated with the proposed algorithm using WorkflowSim and results are compared with state-of-art algorithms. The results indicate an improvement in the performance by minimizing the makespan.