Authors :

Sura Khalil Abd

Address :

Department of Computer Engineering, Dijlah University College, Baghdad, Iraq

Abstract :

Industrial applications widely utilize the Internet of Things (IoT) to create a smart industry to leverage the processing task. The industrial application has Peer-to-Peer (P2P) systems planned to perform resource sharing and allocation in a distributed manner for stagnancy-less task completion. The lengthy task requires several resources to enhance the stagnancy during this process. This paper proposes Peer-dependent Scheduling and Allocation Scheme (PSAS) to address this issue. The PSAS method ensures the distributed and ordered processing of tasks in the smart industry based on their length and resource availability. In the resource allocation, sharing and deadline-based features of the resources are analyzed for task computations. The method of allocation and sharing is independent as guided by predictive learning. This predictive learning lists the possibilities of task completion and shared resource availability. The peer systems assign and process the tasks based on the predictive recommendations. The peer systems interact with the IoT platform to exploit the changes in predictive computation. This learning process is unanimous, regardless of task density and resource availability. The introduced PSAS method efficiency was explored using processing time, task processing ratio, wait time, and stagnancy factor.

Keywords :

Industrial applications, Internet of Things, Peer to Peer, Resource Allocation, Smart Industry, Task Scheduling.