Big Data Analytics in Industrial IoT Using a Concentric Computing Model

• Main Authors: Rehman, Muhammad Habib ur, Ahmed, Ejaz, Yaqoob, Ibrar, Hashem, Ibrahim Abaker Targio, Imran, Muhammad, Ahmad, Shafiq
• Format: Article
• Published: Institute of Electrical and Electronics Engineers (IEEE) 2018
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://eprints.um.edu.my/20898/; https://doi.org/10.1109/MCOM.2018.1700632

The unprecedented proliferation of miniaturized sensors and intelligent communication, computing, and control technologies have paved the way for the development of the Industrial Internet of Things. The IIoT incorporates machine learning and massively parallel distributed systems such as clouds, clusters, and grids for big data storage, processing, and analytics. In IIoT, end devices continuously generate and transmit data streams, resulting in increased network traffic between device-cloud communication. Moreover, it increases in-network data transmissions. requiring additional efforts for big data processing, management, and analytics. To cope with these engendered issues, this article first introduces a novel concentric computing model (CCM) paradigm composed of sensing systems, outer and inner gateway processors, and central processors (outer and inner) for the deployment of big data analytics applications in IIoT. Second, we investigate, highlight, and report recent research efforts directed at the IIoT paradigm with respect to big data analytics. Third, we identify and discuss indispensable challenges that remain to be addressed for employing CCM in the IIoT paradigm. Lastly, we provide several future research directions (e.g., real-Time data analytics, data integration, transmission of meaningful data, edge analytics, real-Time fusion of streaming data, and security and privacy).