QoS-SLA-aware Optimization Framework for IoT-Service Placement in Integrated Fog-Cloud Computing

Fog computing has emerged as a main approach to enhancing cloud services at the network edge, significantly improving the performance of various Internet of Things (IoT) applications. However, the availability of resources in computational devices can fluctuate, affecting application execution within the fog environment. Effective service placement is crucial for service providers to manage multiple real-time IoT applications in fog networks, but previous studies have primarily overlooked minimizing Service Level Agreement (SLA) violations. This paper presents a QoS-SLA-aware optimization framework for IoT service placement in integrated fog-cloud computing. The framework begins with a resource reservation strategy based on the current status of fog devices and historical data to minimize SLA violations. An Improved multi-objective Wild Horse Optimization (I_WHO) algorithm is proposed, utilizing the Sobol Sequence method for optimal IoT service placement. The performance of the I_WHO evolutionary algorithm was evaluated using standard benchmark functions and further assessed through practical simulations with the iFogSim tool. Results demonstrate that the proposed approach enhances service time by up to 31%, reduces energy consumption by up to 18%, lowers service costs by up to 9%, and decreases SLA violations by up to 16% compared to the best existing methods.