Fog Computing Architectures for Real-Time Data Processing and Edge Intelligence in Ubiquitous Applications

The growth of Internet of Things (IoT) devices in the areas of ubiquitous applications has caused a multiplied and diverse data explosion, becoming massive and time-sensitive. Conventional centralized cloud computing models, although providing immense storage capacity and computing capabilities, are becoming inadequate in latency-sensitive application processing because of the lengthy distance of communication, which causes a large latency and congestion in networks. This gap is especially troublesome to real-time applications, including autonomous vehicles and e-healthcare systems, where real-time decision making is of utmost importance to prevent failure of critical proportions. One of the possible solutions to these challenges is to introduce Fog Computing as an important paradigm, which can take the functions of cloud even to the edge of the network and, thus, to serve as an essential mediator between the IoT-based devices and the distant cloud. The paper investigates the system and effectiveness of Fog Computing Architectures that particularly accommodate Real-Time Data Processing and implementation of Edge Intelligence. On-the-fly data processing and analysis at the network edge is made possible by fog nodes, which are distributed geographically with a localized computation and storage capability. This has a substantial effect on network usage and latency, decreasing availability and response time respectively since an entirely cloud-based model would not provide those. Moreover, the use of the latest technologies, such as Artificial Intelligence and Machine Learning, to be integrated into the fog nodes enables the complex, distributed data analytics and smart decisions to be made at the point of data creation. We mention the architectural models and their intrinsic nature like mobility support, geo-distribution and location awareness which are essential to dynamic ubiquitous environments. Finally, the paper addresses the open research issues such as security, privacy, and efficient use of resources, that should be addressed to unlock the full potential of the concept of the use of the fog computing in the industrial and smart city applications.