Entropy based swarm intelligent searching for scheduling deadline constrained workflows in hybrid cloud

In this paper, we consider the problem of scheduling tasks of workflows to heterogeneous VMs (virtual machine) in a hybrid cloud. Workflow scheduling in a hybrid cloud is complex due to factors such as varying electricity prices, uncertain resource requirements, and workflow-specific budget and deadline constraints. To address these challenges, we propose a novel intelligent searching algorithm framework, SEPSO (Swarm Entropy based Particle Swarm Optimization). SEPSO enhances the scheduling process by introducing a swarm entropy, a measure that takes into account the diversification of each iteration. Additionally, we develop rules for sorting workflows and tasks that consider the budget and deadline constraints of each workflow. An iteration-varying flight parameter mechanism is also introduced to balance intensification and diversification during the search process. The components and parameters of our proposed algorithm were statistically calibrated using a comprehensive set of random instances. We then compared SEPSO to adapted existing algorithms for similar problems. Our experimental results demonstrate that SEPSO is effective for the considered problem, offering a solution to workflow scheduling that is more time efficient than traditional particle swarm optimization (PSO) algorithms.