Uncertainty-Aware Scheduling of Hierarchical Realtime Workflows with Deadline Constraints in Cloud Environments

As cloud computing adoption accelerates, scheduling multiple workflows with diverse priorities becomes increasingly challenging. High-priority workflows require strict deadline adherence, while low-priority workflows emphasize cost efficiency. Existing algorithms often overlook priority-specific requirements and uncertainties in task execution and network delays, leading to inefficient resource allocation This paper presents the Real-Time Hierarchical Workflow Scheduling (RHWS) framework, which dynamically allocates resources across workflows with different priorities. RHWS operates in three stages-workflow preprocessing, task allocation, and feedback adjustment-and employs a heuristic algorithm to balance cost and deadline constraints. Experimental results show that RHWS improves service satisfaction by 43.8% and reduces cost deviation by 24.5% compared to existing methods.