Abstract
The industrial Internet of Things (IIoT) enables the interconnection of machines, devices, resources, and computing technologies to improve the reliability of manufacturing services. The role of Software-Defined Networks (SDNs) and Network Function Virtualization (NFV) are exploited in the IIoT environment to ensure effective management and computing resource utilization. Based on the SDN and NFV paradigms, this article introduces a novel elastic computing resource virtualization (ECRV) method to improve the flexibility of resource management in the IIoT. The need for virtualization is obtained by identifying the control and process platforms used in industrial task management. Support vector machine-based classification learning is used to achieve balanced identification, and prevents unnecessary distribution of limited resources, Support vector machine helps to retain flexibility in task control processes that use available industrial resources. By separating the process and control platforms, service dissemination is improved and backlogs in task processing are decreased. The proposed method could provide flexible virtualization and reduces the service response time and task failure.