Intelligent task allocation in edge computing with grey wolf optimization for iot-based applications

With the proliferation of IoT devices, traditional cloud data centers are unable to handle the growing data volumes, compromising support for latency sensitive IoT applications. Cloud-fog computing extends cloud resources to the network edge, offering a potential solution. However, task scheduling across this cloud-fog environment introduces new challenges. This thesis presents a semi-dynamic real-time task scheduling algorithm tailored for the cloud-fog environment. The algorithm efficiently allocates tasks to minimize energy consumption, cost, and makespan. Task allocation is optimized using a modified grey wolf optimizer based on task length, resource requirements, and execution time. Compared to existing methods like genetic algorithms, the proposed algorithm demonstrates superior performance in terms of makespan, cost, and energy usage. The algorithm's ability to optimize task scheduling in the cloud-fog environment enables effective support IoT applications and contributes a novel scheduling approach for emerging cloud-fog computing paradigms.