An efficient task offloading approach to reduce energy consumption in fog architecture

Fog computing architecture is preferable to cloud architecture for resource availability and communication. It provides efficient storage and data processing services that are accessible from the edge of the Internet. However, since most fog setups are battery-operated, there is a high risk of processing failures in applications. Such failures can delay application request processing and prevent them from meeting time-sensitive deadlines. To mitigate this, applications can be run in a power-conscious manner to prevent processing failures caused by power outages. Task offloading refers to mapping tasks to corresponding virtual machines (VMs) to reduce overall energy consumption while ensuring optimal resource utilization. In this paper, we propose a mechanistic optimization technique called GA-COA, which combines a genetic algorithm with the crayfish optimization algorithm to offload tasks in an energy-efficient manner. We have simulated the model and demonstrated the effectiveness of our proposed approach through a comparative analysis with existing techniques. Our results show a remarkable improvement in performance compared to other methods.