Digital twin intelligent system for industrial internet of things-based big data management and analysis in cloud environments

This study surveyed multiple open challenges and problems in the field of sustainable industrial IoTbased big data management and analysis in cloud environments, particularly, the aspects and challenges arising from the fields of machine learning scenarios of cloud infrastructures, artificial intelligence techniques of industrial IoT-based BD analytics in cloud environments, and federated learning cloud systems. Considering that reinforcement learning is a novel technique that allows large data centers such as cloud data centers to influence a more energy-efficient resource allocation, the authors propose an architecture that attempts to combine the features offered by several cloud providers to emerge and achieve an energy-efficient industrial IoT-based big data management framework (EEIBDM) established outside of every user in the cloud environment. As a result, the major goal of this study is the formulation of various aspects of the resource allocation issue, considered from the reinforcement learning scenario, while interacting with the cloud environment to achieve an optimal decision. To achieve this, the authors propose an algorithm for delivering the energy consumption of the CPU through the evaluation of the EEIBDM framework.

As a case study for the future, the authors plan to incorporate security and privacy aspects into proposed system framework to achieve an energy efficient and secure cloud-based management and analysis environment based on industrial IoT, with the help of innovative techniques of reinforcement and federated learning. Thus, this proposed framework could be used in places such as hospitals, schools, and repositories of legal cases to have a more secure environment, in addition to the most energy-efficient environment. These are future directions that extend the authors proposal and plan to be investigated in future research.