On the Scheduling of Industrial IoT Tasks in a Fog Computing Environment
In the industrial sector, a growing number of companies have an ongoing smart factory initiative. In such initiative, previously disparate systems and equipment become connected so the data streams they generate can be turned into actionable insights. Ind
- PDF / 725,559 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 92 Downloads / 226 Views
Internet of Things (IoT) is one of the new disruptive technologies of our time that could profoundly transform the industry ecosystem enabling the concept of smart factories to become a reality. It could potentially change the way industrial facilities and maintenance staff collect and use data to improve the delivery of robot operation. In modern industrial facilities, massive amounts of data originate from a variety of sensors and IoT devices, which monitor in real-time the operations of various systems as diverse as industrial equipment, industrial robots, supply chains, and industrial buildings. c The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 R. Silhavy et al. (Eds.): CoMeSySo 2020, AISC 1294, pp. 966–978, 2020. https://doi.org/10.1007/978-3-030-63322-6_83
On the Scheduling of Industrial IoT Tasks in a Fog Computing Environment
967
By processing IIoT data streams, it would be possible to identify the most significant events and patterns and make appropriate decisions on them in near real-time. It would allow industrial stakeholders to respond to urgent situations with both speed and precision. The data processing chain may involve operations like filtering, aggregating and ultimately storing resulting data for further processing by data analytics applications. Moreover, the ability to federate and process industrial sensor data streams would permit to provide better maintenance of industrial equipment, harness the governance of industrial facilities, and reduce costs. These potential benefits of using IIoT solutions in industrial systems come with challenges with regards to storing, disseminating and processing the vast amounts of sensed data. Many industrial applications use the power and elasticity of the cloud for data storage and processing [2,22]. However, time-sensitive industrial applications cannot bear transmitting data streams to cloud servers for processing because of unacceptable high latency and high network bandwidth requirements. Instead of conveying data to cloud servers for processing and storage, end devices and sensors should pass the data to an edge computing device to aggregate, process, or analyze that data to minimize costs and lower latency while controlling network bandwidth. A substantial benefit of this operation is the reduction of data that must be transmitted and stored in the cloud. IIoT applications in a smart factory are typically organized as workflows of dependent tasks. In order to achieve the expected performance in terms of latency by these applications, it is crucial to schedule the execution of individual tasks of the workflow efficiently in a local fog computing environment. Scheduling of workflows has been investigated in the context of cloud computing and general distributed systems. Our goal is to assess the performance of some well-known scheduling algorithms in terms of average execution time in the context of an industrial fog computing environment. The rest of this paper is organized as follows. Section
Data Loading...
