A queuing theory model for fog computing

Abstract

Under many scenarios where resources may be scarce or a good Quality of Service is a requirement, appropriately sizing components and devices is one of the main challenges. New scenarios, such as IoT, mobile cloud computing, mobile edge computing or fog computing, have emerged recently. The ability to design, model and simulate those infrastructures is critical to dimension them correctly. Queuing theory models provide a good approach to understanding how a given architecture would behave for a given set of parameters, thus helping to detect possible bottlenecks and performance issues in advance. This work presents a fog-computing modelling framework based on queuing theory. The proposed framework was used to simulate a given scenario allowing the possibility of adjusting the system by means of user-defined parameters. The results show that the proposed model is a good tool for designing optimal fog architectures regarding QoS requirements. It can also be used to fine-tune the designs to detect possible bottlenecks or improve the performance parameters of the overall environment.

Model of optimal distribution of network resources with constraints on quality of service indicators

“…These models in [5], [6] do not take into account the utilization of the router interface queues by their length. A model of dynamic guaranteed queuing is considered [14]- [19]. Furthermore, packet delay and buffer size are evaluated.…”

Key Challenges and Emerging Technologies in Industrial IoT Architectures: A Review

“…We have grouped the widely used emerging technologies in the literature and focused on evaluating the scope of each technology in addressing the main IIoT requirements in those architectures. The following are some of the emerging technologies we have observed in developing the IIoT layered architectures: Edge/Fog Computing: Fog computing brings the cloud services closer to the ground mobile devices to offload the processing burden, improve the Quality of Service (QoS) of a system, and save resources [ 68 ]. Based on the information given by the National Institute of Standards and Technology (NIST), the fog computing should have the main characteristics of supporting the geographical distribution, low latency, interoperability features, scalability, and real-time interactions rather than batch processing [ 69 ].…”

“…Edge/Fog Computing: Fog computing brings the cloud services closer to the ground mobile devices to offload the processing burden, improve the Quality of Service (QoS) of a system, and save resources [ 68 ]. Based on the information given by the National Institute of Standards and Technology (NIST), the fog computing should have the main characteristics of supporting the geographical distribution, low latency, interoperability features, scalability, and real-time interactions rather than batch processing [ 69 ].…”

QoE Aware Application Placement in Fog Environment Using SAW Game Theory Method

“…Keshavarznejadet et al [20] use optimization methods to discharge tasks in the form of a multiobjective optimization problem that focuses on reducing the power consumption of the entire system and delaying the execution of tasks. Mas et al [21] present a fog computing modeling framework in which, using user-defined parameters, the optimal system is set according to QoS requirements. Nardin et al [22] use a Cloud-based microservice application to reduce energy consumption and allocate the most appropriate source to an application.…”