QoS Based Optimal Resource Allocation and Workload Balancing for Fog Enabled IoT

This paper is aimed at efficiently distributing workload between the Fog Layer and the Cloud Network and then optimizing resource allocation in cloud networks to ensure better utilization and quick response time of the resources available to the end user. We have employed a Dead-line aware scheme to migrate the data between cloud and Fog networks based on data profiling and then used K-Means clustering and Service-request prediction model to allocate the resources efficiently to all requests. To substantiate our model, we have used iFogSim, which is an extension of the CloudSim simulator. The results clearly show that when an optimized network is used the Quality of Service parameters exhibit better efficiency and output.