Elsevier

Computer Networks

Volume 187, 14 March 2021, 107827
Computer Networks

Independent and tailored network-slicing architecture for leveraging industrial internet of things job processing

https://doi.org/10.1016/j.comnet.2021.107827Get rights and content

Abstract

Industrial automation and management designed using the Internet of Things (IoT) paradigm leverage the functional and reliable operations of the production industry. Different network slices for independent and collaborative functioning connect the internal operations between different industrial IoT (IIoT) layers. This article proposes an independently-tailored network-slicing architecture for improving the mutual operations of different layers of IIoT functions. The proposed architecture helps improve the processing of scheduled jobs in different layers in an associated manner. The association process between the scheduled processes is independently analyzed for improving the swiftness in IIoT production outcomes. The scheduling and network-slicing features are recommended based on the processing and production outcomes of the analyzed industry using regression learning, which helps to assign associated processes in a queue depending on the time and production unit availability. Therefore, the architecture's network-slicing feature is modified based on the recommendations from the learning in providing seamless processing and interconnection support for IIoT production enhancements. The proposed architecture's performance is assessed using the metrics production responses, processing rate, processing time, process lag, and resource allocations. The proposed architecture achieves 10.35%, 15.32%, and 9.05% high response ratio, processing rate, and resource allocation. It reduces processing time and lag factor by 9.25% and 9.19% respectively. This observation is with respect to the processing units.

Introduction

Industrial automation has been improved by assimilating the Internet of Things (IoT) paradigm. Industrial IoT (IIoT) helps us capture and turn trustworthy and secure data in real-time into useful information for companies. It helps reduce resource usage to increase our competitiveness and performance. By using IIoT sensors in equipment creates maintenance warnings based on conditions. The manufacturers can save resources, minimize costs, remove downtime, and improve operational efficiency by securing the prescribed working environment for machinery. The complete set of heterogeneous devices, computing technologies, and other networking infrastructures are inherited for machine and control functions of the industry [1]. An automated industry consists of interconnecting processing, control, monitoring, planning, and management layers for cooperative task allocation and processing [2]. The IoT resource allocation, sharing, and processing features are administered between the layers to improve industrial production performance [3]. Specifically, job allocation and processing are managed between processing, control, and monitoring layers to enhance agreements in job response delivery [4]. The request for allocating jobs is processed at the initial stage from the IoT connected layers distributed depending on resource availability. For processing and response delivery, the IoT environment's distributed computing nature and artificial intelligence are jointly exploited in industrial automation to help meet the consumer's requirements by achieving a better quality of service (QoS) [2, 5].

The jobs managed in an industrial environment vary with the time and type of the performed process. The available processing units/machines should also improve the industry's service responses [6]. In particular, the job allocation and processing rate of available machines determine the industrial automation system's performance. Therefore, job scheduling and resource allocation are critical factors in controlling industry operations [7]. Job scheduling and resource allocation are administered using IoT computing techniques in the processing and control layers that are periodically verified by the monitoring layer [8]. The inherited IoT computing paradigms as external support for the IIoT Job streamlining perform idle machine identification, resource lag, queuing jobs, and rescheduling. Heuristic rules are chosen from a set of workstations on a specific computer without previously evaluating the choices’ results. As the most popular planning approach, several dispatching rules are recognized for a range of planning requirements. For example, some of these purposes involve dispatching rules to minimize completion or execution times, optimize standard deviations, or generate just-in-time.

Optimal resource allocation and scheduling help optimize the flow of jobs and responses by regularly engaging machine processing. Such a seamless integration and assistance from computing techniques in the IIoT environment prevent uncontrolled processing delay and response lags [9,10].

Another typical IIoT performance-influencing factor is resource availability and allocation. Both computing and processing unit resources are significant for improving the industry's response rate by controlling process lags [11]. Resource availability is subject to using and accepting tasks. The connected network's availability is essential in all industrial layer functions [12]. Therefore, network-slicing concepts are assimilated into the IoT platform. Network slicing ensures the availability of resources by virtualizing and multiplexing them for any range of requests. The network slice-processing functions as an independent resource architecture [13]. The network division's idea is to use network infrastructures to create several sub-networks for services and applications by splitting each subnet to create an individual network for its operation, splitting the physical network resources. It consists of a series of network functions that implement well-defined actions and interfaces. Multiple network tasks will be channeled through the virtual network infrastructure to create an end-to-end network instance representing the network features requested by the service.

The slices provide distributed computing and processing requirements. With the help of sliced network resources, the migration of overflowing and waiting jobs is processed on time [14]. The computing and decision-making systems inducted in the IoT platform efficiently determine the task migration. The smart decision-making and intelligent computing techniques in IIoT aim to improve the industrial environment's response to different real-time applications [15]. The contributions of the article are summarized below.

  • Designing an independent and tailored network-slicing architecture for improving the job-handling rate of IoT-assisted smart industries

  • Designing a prompt scheduling method for managing diverse job flows for processing, mitigating the process lag, and improving the processing rate

  • Performing a tailored virtualization process in an industrial environment for optimal resource use and improve the seamlessness in task processing

  • Performing a comparative analysis for verifying the consistency of the proposed method using different metrics with existing methods

The organization of the article is as follows: Section 2 discusses the related works from the past with its pros and cons. In Section 3, the proposed architecture is discussed with its functions and descriptions, followed by the experimental and comparative study in Section 4. Section 5 concludes the article with a final briefing.

Section snippets

Related works

Jin et al. [16] presented a content-centric cross-layer scheduling solution (CONCISE) to address the inner layer network traffic. This study uses the routing structure and multiple content-based schedulers. Processing this decreases the delay and improves the packet delivery ratio. Yang et al. [17] proposed an optimal geographical placement to address the base station's entire missing probability. The authors overcame this probability issue by implementing a low-complexity near-optimal

Proposed independently-tailored network-slicing architecture (ITNSA)

IIoT integrates smart manufacturing machines, information, and communication technologies using different layers in the industry. This study focuses on improving resource allocation and processing scheduled jobs in various industrial layers in an associated manner. For reliable resource allocation and task processing, ITNSA is developed. In the proposed job-processing architecture, the production unit and scheduling must satisfy the association. Besides, resource allocation is facilitated using

Results and discussion

In this section, the performance of the proposed independently-tailored network-slicing architecture (ITNSA) is discussed. The IIoT environment is modeled using the opportunistic network environment (ONE) simulator [30]. The simulations are performed in a standalone system with 6 GB physical memory and 500 GB storage space. The processing speed of the physical device is 2.4 GHz with 64-bit architecture support. Table 2 presents the in-simulation parameters and their configurations.

In the above

Conclusion

An ITNSA was designed for improving the service responses of an automated IoT-assisted industry. The scheduled and overflowing jobs in the IIoT environment are managed effectively by exploiting the association between different functional layers. For the association and cooperative processing of the service requests, scheduling and network-slicing features are exploited for improving the response rate. In this scheduling and network slicing, regression learning identifies the optimality in

CRediT authorship contribution statement

Zafer Al-Makhadmeh: Conceptualization, Methodology, Software, Writing - original draft. Amr Tolba: Visualization, Investigation, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no conflicts of Interest

Acknowledgments

The authors extend their appreciation to the Deanship of Scientific Research at King Saud University for funding this work through research group No. (RG-1438–027).

Zafer Al-Makhadmeh received the M.Sc. and Ph.D. degrees from the Department of Computer Engineering, Faculty of Information and Computer Engineering, Kharkov National Technical University of Ukraine, in 1998 and 2001, respectively. He is currently an Associate Professor with the Department of Computer Science, Community College, King Saud University, Saudi Arabia. He has authored/coauthored over 30 scientific papers in top ranked (ISI) international journals and conference proceedings. His-main

References (30)

  • S. Kharb et al.

    Fuzzy based priority aware scheduling technique for dense industrial IoT networks

    J. Netw. Compu. Appli.

    (2019)
  • R. Casado-Vara et al.

    IoT network slicing on virtual layers of homogeneous data for improved algorithm operation in smart buildings

    Futu. Gener. Compu. Syst.

    (2020)
  • F. Foukalas

    Cognitive IoT platform for fog computing industrial applications

    Compu. Electr. Eng.

    (2020)
  • P. Zeng et al.

    Time-slotted software-defined Industrial Ethernet for real-time Quality of Service in Industry 4.0

    Future Gener. Comput. Syst.

    (2019)
  • X. Huang

    Quality of service optimization in wireless transmission of industrial Internet of Things for intelligent manufacturing

    Int. J. Adv. Manufact. Technol.

    (2019)
  • Cited by (6)

    • Elastic Computing Resource Virtualization Method for a Service-centric Industrial Internet of Things

      2021, Computer Networks
      Citation Excerpt :

      The sophisticated computing and communication platform of the IIoT paradigm helps meet an industry's task and request demands. In a conventional IIoT environment, resource limitations and device restrictions are mitigated through scalable resource allocation and task processing [4]. In an industrial environment, operations are streamlined by providing distributed and pervasive allocations and computing strategies.

    • Overview on intelligent scheduling models and methods for industrial Internet-of-things

      2022, Jisuanji Jicheng Zhizao Xitong/Computer Integrated Manufacturing Systems, CIMS
    • Sharing Economy in Energy Markets: Modeling, Analysis and Mechanism Design

      2022, Sharing Economy in Energy Markets: Modeling, Analysis and Mechanism Design

    Zafer Al-Makhadmeh received the M.Sc. and Ph.D. degrees from the Department of Computer Engineering, Faculty of Information and Computer Engineering, Kharkov National Technical University of Ukraine, in 1998 and 2001, respectively. He is currently an Associate Professor with the Department of Computer Science, Community College, King Saud University, Saudi Arabia. He has authored/coauthored over 30 scientific papers in top ranked (ISI) international journals and conference proceedings. His-main research interests include cloud computing, image processing, computer vision, and intelligent systems.

    AMR TOLBA received the M.Sc. and Ph.D. degrees from Mathematics and Computer Science Department, faculty of science, Menoufia University, Egypt, in 2002 and 2006, respectively. He is currently an Associate Professor at the Faculty of Science, Menoufia University, Egypt. He is currently on leave from Menoufia University to the Computer Science Department, Community College, King Saud University (KSU), Saudi Arabia. Dr Tolba serves as a technical program committee (TPC) member in several conferences. He has authored/coauthored over 75 scientific papers in top ranked (ISI) international journals and conference proceedings. His-main research interests include socially aware networks, vehicular ad-hoc networks, Internet of Things, intelligent systems, Big Data, recommender systems, and cloud computing.

    View full text