•  
  •  
 

Abstract

The late 1990s saw the rise of the edge computing network paradigm, as well as an increase in the number of IoT de-vices. This concept is viewed as a link between cloud servers and end-devices, bringing processing and storage re-sources closer to clients. As a result of its low latency and high performance, researchers and developers have expressed interest in it. However, this paradigm confronts a number of obstacles and restrictions, including restricted and hetero-geneous resources at network edges. In this paper, we provide a detailed review of heterogeneous resources in edge network infrastructures using a three-dimensional method. These three dimensions in this concept correspond to the edge computer layers, hardware layers, and software layers of the edge network paradigm infrastructure ecosystem. This review considers Artificial Intelligence (AI), which classifies cutting-edge works into two categories: AI-based and non-AI-based solutions based on research issues such as fault tolerance, power consumption, resource utilization, re-source allocation, latency, device ID, clustering, heuristic-based, and meta-heuristic-based. Because reviews in this field rarely address full research concerns linked to this research topic. This review provides a sufficient overview to address the majority of open research questions. We examine and compare outstanding issues in AI-based and non-AI-based systems, focusing on evaluation metrics for meeting Quality of Services (QoS) and Quality of Experience (QoE) stand-ards. We expect that this evaluation will assist developers and researchers in determining the appropriate solution from research issues to achieve their objectives in building IoT technology and edge computing networks.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Download

Share

COinS