Towards an efficient scheduling strategy based on multi-objective optimization in fog environments

Meeting Quality of Service (QoS) requirements is crucial for Internet of Things (IoT) applications, such as smart healthcare, industrial automation, and intelligent transportation, due to their diverse and often critical nature. Meeting QoS requirements is crucial for IoT applications due to their diverse and often critical nature. Ensuring high QoS guarantees that these applications function smoothly and efficiently, leading to enhanced user experiences and system reliability. With the rapid growth of the IoT and the increasing demand for data processing near the source, fog computing environments emerged as an intermediate layer between cloud and edge devices. Hence, robust QoS management is essential for IoT systems' successful deployment and operation. Meanwhile, utilizing computing resources in the cloud-fog ecosystem is increasingly important and requires an efficient workflow scheduling strategy. This paper proposes an efficient Workflow Scheduling strategy based on Multi-objective Optimization considering Pareto front in fog environments (WSMOP) to address this issue. Our strategy addresses the challenges of resource management and workflow scheduling in fog environments by optimizing multiple objectives, including makespan (total time needed to complete all tasks), energy consumption, latency, throughput, and resource utilization. WSMOP uses an advanced meta-heuristic technique named Open-Source Development Model Algorithm (ODMA) for optimization work. We used the CloudSim simulator for performance evaluation, comparing WSMOP against advanced methods, including NSGA-II, AOAM, HDSOS-GOA, PSO-SA, and BAHA-KHA. Extensive simulations and real-world experiments demonstrate the effectiveness and efficiency of our proposed strategy in enhancing overall system performance and meeting QoS demands in fog computing scenarios. Specifically, WSMOP reduces the average makespan and energy consumption by 1.5% and 2.3% compared to the best existing method, respectively.