Optimizing Bandwidth Cost, Energy Consumption, Delay, and Load in a Cooperative Fog Environment

Fog computing is an emerging architecture that provides storage, processing power, and networking capabilities for Internet-of-Things (IoT) devices. Devices using fog computing can assign computationally demanding tasks and data to adjacent fog nodes. This approach can reduce the bandwidth cost, energy consumption, and delay. However, to reduce one cost, another cost might increase, such as reducing energy consumption may increase the delay. Therefore, balancing these costs efficiently is crucial in fog computing. In this paper, we present a joint optimization strategy for bandwidth cost, energy consumption, delay minimization, and load balancing in a cooperative fog environment. Our approach involves using a Mixed-Integer Linear Programming (MILP) model to optimize the total cost and distribute the workload effectively among the fog nodes. We present a comprehensive evaluation of our approach, using simulations to demonstrate its effectiveness in reducing energy consumption, delay, and bandwidth cost. Our results demonstrate that our method can notably enhance the performance of fog computing systems, making them more efficient and cost-effective for Internet of Things applications.