Digital twin intelligent system for industrial internet of things-based big data management and analysis in cloud environments
Christos Stergiou, Kostas E. Psannis
Abstract
This work initially surveys and illustrates the multiple open challenges in the field of industrial IoT-based Big Data management and analysis in Cloud environments. Challenges arise from fields of Machine Learning in the Cloud infrastructures, A.I. techniques of Big Data Analytics in the Cloud environments, and Federated Learning Cloud systems try to be clarified. Additionally, Reinforcement Learning is a novel technique that allows large data centers such as Cloud data centers to affect a more energy-efficient resource allocation. Moreover, we propose an architecture that tries 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 Cloud. IoT data could be integrated with techniques such as Reinforcement and Federated Learning to achieve a Digital Twin scenario, for the virtual representation of industrial IoT-based Big Data of machines and rooms temperatures. Furthermore, we propose an algorithm for delivering the energy consumption of the infrastructure through the evaluation of the EEIBDM framework. Finally, some future directions as an expansion of our research are illustrated.