Reliable and Cost-Efficient Workflow Scheduling in Edge-Cloud Environment

With the growing demand for real-time and energy-efficient computation, edge computing has emerged as a viable alternative to traditional cloud computing, offering reduced latency and faster task execution.However, existing workflow scheduling (WS) techniques in edge-cloud environments primarily focus on minimizing energy consumption and meeting task deadlines, while cost optimization remains largely underexplored.To address this gap, this work proposes a novel Reliable and Cost-Efficient Workflow Scheduling (RCE-WS) approach that combines intelligent task partitioning with soft computing-based optimization.The novelty of RCE-WS lies in its dualobjective design: minimizing execution cost and enhancing reliability by strategically distributing workflow tasks across edge and cloud resources.RCE-WS was evaluated using two widely recognized scientific workflow datasets-Epigenomics and SIPHT-under a CloudSim-based simulation environment.Comparative analysis against four existing approaches, Resource-Prediction Scheduling-Error Optimization (RPSEO), Energy-Aware Reliability-Enhancement Scheduling (EARES), Priority-Adjustment Critical-Priority Heterogeneous-Earliest Finish-Time (PACP-HEFT), and Scheduling Workflow-Partial-Critical-Paths (SWPCP) shows that RCE-WS achieves an average makespan reduction of 15.94% and 70.22%, average cost reduction of 23.24% and 43.79%, and reliability improvements of 1.662% and 2.001%, for Epigenomics and SIPHT respectively, when compared with existing approaches.These results confirm the effectiveness of RCE-WS in delivering cost-efficient and reliable workflow scheduling in dynamic edge-cloud infrastructures.