Optimizing Cloud-Fog Workloads: A Budget Aware Dynamic Scheduling Solution

With the rise of demanding computational applications, individuals and businesses reap significant benefits. However, processing power limitations on local devices hinder further progress. These local resources also require substantial financial and human investment. Fortunately, the adoption of cloud computing (CC) platforms alleviates these concerns. CC enables offloading computationally intensive tasks to the cloud, reserving local resources for simpler jobs. However, as CC operates on a pay-as-you-go model, finding an approach that minimizes both execution time and cost is crucial. While research efforts have attempted to address this challenge, solutions remain limited. This paper proposes a novel architecture leveraging cloud provider resources and local computing power on fog computing devices. The core of this framework is a dynamic task scheduling solution that minimizes completion time within the cloud service while considering network conditions and customer costs for service usage. Simulations and comparisons with existing scheduling approaches demonstrate that our proposed method delivers robust performance with significant cost savings for cloud users.