Improved Political Optimizer and Deep Neural Network-based Resource Management Strategy for fog Enabled Cloud Computing

Cloud-based processing of Internet of Things (IoT) applications isn't always the best option, especially when it comes to time-sensitive applications. For managing the massive amounts of data bandwidth required by end devices, fog and edge computing offer a viable alternative. Large volumes of newly generated data must be processed locally, rather than in the cloud, according to these paradigms. Allocating resources, balancing workloads, providing resources, scheduling tasks, and ensuring quality of service are all aspects of resource management that play an important role in cloud-based IoT systems with the goal of improving system performance. In this paper, we discuss different approaches to resource management in the context of cloud, fog, and edge computing. We introduce a methodology Improved Political optimizer (IPO) with Deep Neural Networks (DNN) for ranking resource management algorithms in the context of cloud, fog, and edge systems. Therefore, we start with the challenges of resource management in that industry. Current research contributions can be better categorised, which will aid in the creation of an evaluation scheme. There is a substantial addition made by reviewing and analysing research papers on methods for resource management. Existing Fog resource managers only account for a subset of Fog resource management criteria, such as system network bandwidth and Response Time. The administration of computing assets in the cloud, fog, and at the network's periphery are the primary topics of our studies. The purpose of this analysis is to develop a set of criteria for judging algorithms used for managing resources in cloud, fog, and edge environments. We employed Deep Neural Networks and the Improved Political Optimizer in our search for better methods of managing available resources (DNN).