Fog computing bridges the gap between IoT devices and cloud servers by providing low-latency computational resources closer to the network edge. Despite its potential, the rapid increase in IoT applications with diverse resource and quality-of-service (QoS) requirements presents significant challenges in application deployment and resource optimization. This paper addresses these challenges by introducing a comprehensive application placement framework designed to optimize execution time and energy consumption in a heterogeneous fog environment. The proposed framework consists of three phases. A pre-scheduling method is developed to efficiently allocate tasks by analyzing workflows to reduce computation delays and energy usage. Leveraging an Improved Memetic Algorithm (IMA), this strategy enables effective scheduling of parallel IoT workflows across fog and cloud servers, ensuring balanced resource utilization and enhanced scalability. A lightweight recovery method is incorporated to address runtime failures, ensuring the robustness and reliability of task execution. The performance of the proposed framework is evaluated using real and synthetic IoT workflows in the iFogSim environment. Experimental results demonstrate that the framework achieves a 65% reduction in the weighted cost and a 51% decrease in execution time compared to existing approaches. This makes it a promising solution for managing resource-intensive IoT applications in fog computing environments.

Fog computing bridges the gap between IoT devices and cloud servers by providing low-latency computational resources closer to the network edge. Despite its potential, the rapid increase in IoT applications with diverse resource and quality-of-service (QoS) requirements presents significant challenges in application deployment and resource optimization. This paper addresses these challenges by introducing a comprehensive application placement framework designed to optimize execution time and energy consumption in a heterogeneous fog environment. The proposed framework consists of three phases. A pre-scheduling method is developed to efficiently allocate tasks by analyzing workflows to reduce computation delays and energy usage. Leveraging an Improved Memetic Algorithm (IMA), this strategy enables effective scheduling of parallel IoT workflows across fog and cloud servers, ensuring balanced resource utilization and enhanced scalability. A lightweight recovery method is incorporated to address runtime failures, ensuring the robustness and reliability of task execution. The performance of the proposed framework is evaluated using real and synthetic IoT workflows in the iFogSim environment. Experimental results demonstrate that the framework achieves a 65% reduction in the weighted cost and a 51% decrease in execution time compared to existing approaches. This makes it a promising solution for managing resource-intensive IoT applications in fog computing environments.