2019年边缘计算技术发展研究

A Survey on Edge Computing in 2019

边缘计算将数据处理任务放置在网络边缘(靠近数据源)的设备上,可以快速实现实时数据处理,具有低延迟、隐私安全性和灵活性的特点,被认为是“人工智能的最后一公里”。近年来,边缘计算的飞速发展在不断推动边缘人工智能物联网的发展。对2019年边缘计算的相关研究与成果进行了综合评述,并对未来发展趋势进行了展望。首先,梳理2019年面向边缘计算的人工智能软硬件进展,科技巨头纷纷推进战略部署,市场火爆;然后,总结2019年在降低功耗方面的最新成果,指出降低功耗是边缘计算技术落地的关键点;随后,从嵌入超低功耗人工智能芯片和异构计算架构两个方向概述边缘计算最新发展趋势;最后,从电源功耗、人工智能芯片、安全隐私、5G等四个角度讨论边缘计算面临的困难与挑战。

Edge computing places data processing tasks on the devices which near the edge of the network,close to the data source.This computing methodology can realize real-time data processing very quickly,with characteristics of low latency,privacy security and flexibility,which is considered as the last kilo-meter of artificial intelligence(AI).In recent years,the rapid development of edge computing is constantly promoting the development of edge AI internet of things(IoT).This paper reviews the related research and achievements of edge computing in 2019,and looks forward to the future development trend.First of all,this paper analyzes the recent development of AI software and hardware for edge computing in 2019,wherein technology giants have promoted strategic deployment which causes the market’s booming.Secondly,this paper summarizes the latest achievements in reducing power-consumption in 2019,and shows that reducing power-consumption is the key point for the implementation of edge computing technology.Thirdly,the latest development trend of edge computing is overviewed from two directions of embedded power-consumption power AI chip and heterogeneous computing architecture.Finally,the problems faced by edge computing about power-consumption,AI chip,security privacy and 5G,are discussed.