Workflow Scheduling Algorithm Based on Budget Constraints in a Cloudy Environment

In addressing the conflict between workflow execution time and cost in a multi-cloud environment with budget constraints, and considering the diverse billing cycles in such an environment, a novel Ant Colony Optimization (ACO) algorithm, named Multi-Cloud Ant Colony Optimization (MACO). The algorithm utilizes the ACO algorithm to prioritize workflow tasks. During the resource allocation phase, a greedy strategy is employed to rent resources, dynamically adjusting the types and quantities of rented resources to meet user budget constraints. Additionally, the algorithm introduces a novel task priority calculation method as heuristic information for the MACO algorithm. Experimental comparisons with NBWS, PBHEFT, and StreamWS algorithms show that, under budget constraints, MACO demonstrates the ability to significantly improve scheduling success rates and reduce workflow execution time. The experimental results demonstrate that the algorithm effectively improves the success rate of workflow execution and reduces workflow execution time in a multi-cloud environment with budget constraints, successfully addressing the time optimization problem in workflow scheduling under budget constraints.