Mobile Edge Computing: Architecture, Use-cases, Applications
Résumé
By enormous growth in IoT and smart devices and the advent of many new applications, Internet traffic volume has been growing exponentially. Analyzing such flooding traffic requires enormous compute and bandwidth and raises privacy concerns. Edge platforms can become the tool to ease the burden by bringing resources to the proximity of data. Therefore, new architectures, which bring network functions and contents to the network edge, are proposed, i.e., mobile edge computing and caching. In this survey, we make an exhaustive review on the literature research efforts on mobile edge networks. We give an overview of mobile edge networks, including definition, architecture , and application and use-cases. We then survey the issues related to computing, caching, and communication techniques at the network edge with the focus on applications and use cases of mobile edge networks. I Introduction Many mobile applications relay on remote data centers. This imposes high loads of data in mobile networks due to uploading and downloading data to and from data centers. Demand of bandwidth is expected to be doubled each year [1]. Moreover, mobile devices have the computation power of a sever in a decade ago. With the increase of computational power, novel mobile applications such as Augmented Reality (AR) become realistic. On the other hand, Internet of Things (IoT) enables resource-limited devices to interconnect with the Internet. To support IoT, new techniques such as computation offloading are introduced in order to offload a part of the computation to a remote cloud server. Although the offloading brings lower energy consumption and computation power, it may introduce extra latency while exchanging data between the device and cloud servers. To address this issue, cloudlet offloading has become prevalent. In this technique the computational task is offloaded to a cloud server in the proximity of the mobile device using Wi-Fi [2]. However, Cloudlet has its own pitfalls. First it is only accessible through Wi-Fi that will only cover a short range, Second, it is not scalable in terms of resource provisioning. To mitigate all the limitations and issues mentioned above, a new paradigm called Mobile Edge Computing (MEC) has been introduced. This concept was firstly proposed by the European Telecommunications Standard Institute (ETSI) 1
Origine : Fichiers produits par l'(les) auteur(s)