Mobility-aware hierarchical fog computing framework for Industrial Internet of Things (IIoT)

Abstract

The Industrial Internet of Things (IIoTs) is an emerging area that forms the collaborative environment for devices to share resources. In IIoT, many sensors, actuators, and other devices are used to improve industrial efficiency. As most of the devices are mobile; therefore, the impact of mobility can be seen in terms of low-device utilization. Thus, most of the time, the available resources are underutilized. Therefore, the inception of the fog computing model in IIoT has reduced the communication delay in executing complex tasks. However, it is not feasible to cover the entire region through fog nodes; therefore, fog node selection and placement is still the challenging task. This paper proposes a multi-level hierarchical fog node deployment model for the industrial environment. Moreover, the scheme utilized the IoT devices as a fog node; however, the selection depends on energy, path/location, network properties, storage, and available computing resources. Therefore, the scheme used the location-aware module before engaging the device for task computation. The framework is evaluated in terms of memory, CPU, scalability, and system efficiency; also compared with the existing approach in terms of task acceptance rate. The scheme is compared with xFogSim framework that is capable to handle workload upto 1000 devices. However, the task acceptance ratio is higher in the proposed framework due to its multi-tier model. The workload acceptance ratio is 85% reported with 3000 devices; whereas, in xFogsim the ratio is reduced to approx. 68%. The primary reason for high workload acceptation is that the proposed solution utilizes the unused resources of the user devices for computations.

A Dynamic IIoT Framework Based on the Publish–Subscribe Paradigm

“…The solutions for fog computing from the specialized literature that are analyzed in different review or survey papers [ 13 , 17 , 18 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ] specify that either they are presented at an abstract level or they are solutions for specific applications. An essential advantage of the proposed solution, which distinguishes it from the solutions presented in the specialized literature, is customizability, because it allows the instantiation of objects and their interconnection to implement specific solutions.…”

“…The main solutions for fog computing from the specialized literature that are analyzed in different reviews or survey papers [ 13 , 17 , 18 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ] present the high-level parts of the fog nodes, such as security, the task scheduling, or resource scheduling when resources are limited; these papers are not focused on how data are processed within fog computing.…”

Онтологічний Підхід До Перерозподілу Навантаження Інтернету Речей

“…На базі туманної концепції ефективно реалізуються різні розподілені системи, для яких швидкодія є найважливішим критерієм. Такі системи є вкрай затребуваними у багатьох областях людської діяльності, зокрема, у сучасних системах Інтернету речей, які мають високу та надвисоку щільність кінцевих датчиків [4][5][6][7]. Але використання туманної концепції у таких випадках потребує доволі частого перерозподілу навантаження між обчислювальними вузлами.…”

At the Confluence of Artificial Intelligence and Edge Computing in IoT-Based Applications: A Review and New Perspectives

“… Frameworks and simulators: To support real-time analysis and development of fog computing, the authors of [ 141 ] developed modular simulation models for service migration, dynamic distributed cluster formation, and microservice orchestration for edge/fog computing based on real-world datasets. In [ 142 ], the authors proposed a multilayer fog deployment framework for job scheduling and big data processing in an industrial environment. With regard to device computation, we find the following.…”