Edge computing for interactive media and video streaming

Video streaming and computer games are among the most popular and highest bandwidth consuming media in the Internet. Video contents consume around 70% of the total bandwidth usage in the Internet today. Advancements in media generation tools, high processing power, and high speed connectivity have enabled generation of live, interactive, multi-view media generation. Cognitive assisted and online multi-player gaming have unlocked new horizons for gaming experience. However, such interactive gaming, and multi-view and 360-degree view videos, are currently limited by delay intolerance and excessive bandwidth usage. Edge computing is the name of a set of new technologies, such as cloudlets, micro data centers, fog, and mobile edge computing. It aims to provide storage and computational resources near to user at the network edge, to minimize latency and response time. Edge computing is foreseen as a significant enabler of Internet of Every Thing (IoET) era by extending the cloud services and resources at the end of the network to deliver very low latency and real-time communication. It can provide significant services to video and gaming applications and enable new stream of interactive multimedia era. In this paper, we highlight some of the potentials and prospects of edge computing for interactive media, and present some preliminary works in the area. We shed light on how edge computing can be used to tackle various challenges faced by todays interactive media application. We also present the benefits of using edge computing to save cost, bandwidth, and energy in multimedia applications, video streaming, and transcoding.

I. Introduction

In the last decade, various new technologies emerged with an aim to minimize the physical distance and number of hops between the users and service delivery points, such as cloudlets [1], fog [2], micro data centers [3], [4], and mobile edge computing [5], [15], [16]. All of the aforementioned technologies hold one common aspect and common objective, i.e., to bring the cloud resources from far located giant cloud data centers to small scaled facilities situated at the edge of the network, aiming to minimize the response time. We name these multiple technologies as “Edge computing technologies” to discuss the potentials of this new paradigm in terms of interactive media and video streaming. Edge computing extends the storage and computational resources to a very fine grained level to the edges of the networks, where user can access these resources in very less number of hops, resulting in real-time interaction, low latency and instant response, location awareness, and support for mobility, to name a few. It is worthy to note that edge computing technologies are not going to replace the current cloud model, instead the edge technologies will augment and extend the cloud services to the edge of the network.