Author:
Gu Jianhua,Feng Jianhua,Xu Huiyang,Zhou Ting
Abstract
With the explosive growth of the demand for computing power in the era of digital economy and the continuous enhancement of the computing power of terminals, how to provide high bandwidth, low-latency, and low-cost service by leveraging the user devices’ computing, storage, and network resources has become a research hotspot. However, due to differences in magnitude, architecture, and performance, the existing technologies for cloud computing and edge computing need to overcome many challenges such as android-based devices not supporting container technologies. In this paper, a terminal-side computing force network (TSCFN) architecture is proposed, which realizes the unified computing power management of massive user devices by layered and distributed architecture with highly dynamic and domain-federated deployments. At the same time, we propose a cloud-native container resource scheduling scheme based on the Android system to enhance the scalability of TSCFN. Taking the CDN service as a use case, the experiment results show that services provided by TSCFN can reduce latency and improve resource utilization, especially in an unstable network status. Compared with traditional CDN, delay duration of HomeCDN based on TSCFN is decreased by 96% in a bad network environment.
Subject
Electrical and Electronic Engineering,Computer Networks and Communications,Hardware and Architecture,Signal Processing,Control and Systems Engineering
Reference26 articles.
1. Computing network: A new multi-access edge computing;Lei;Telecommun. Sci.,2019
2. White Paper on Computing Power Aware Network Technology,2019
3. Computing Force Network White Paper,2021
4. White Paper on Computing Network Architecture and Technology System;Institute,2020
5. Cloud Convergence 2030 Technology White Paper;Corporation,2020
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献