Machine Learning Based Task Distribution in Heterogeneous Fog-Cloud Environments

In order to improve the quality of service for delay-sensitive applications, in this paper, we propose Machine Learning based Task Distribution (MLTD) technique, which utilizes Artificial Neural Networks to distribute the tasks between the fog and cloud resources intelligently. This technique takes the diversity of servers (in terms of computing power) in addition to their workloads at the time of task distribution into account for providing the best possible response time. Evaluating the performance of our proposed technique, we utilized it in a real-world testbed and investigated its performance in different conditions. In comparison with similar methods, the achieved results show that MLTD improves the response time when the workloads of servers change continuously and reduces the internet bandwidth utilization in most cases.