A Transparent and User-Centric Approach to Unify Resource Management and Code Scheduling of Local,Edge,and Cloud  

作　　者：ZHOU Yuezhi ZHANG Di ZHANG Yaoxue 

机构地区：[1]Tsinghua University [2]Central South University

出　　处：《ZTE Communications》2017年第4期3-11,共9页中兴通讯技术（英文版）

基　　金：supported in part by Initiative Scientific Research Program in Tsinghua University under Grant No.20161080066;in part by International Science&Technology Cooperation Program of China under Grant No.2013DFB10070

摘　　要：Recently, several novel computing paradigms are proposed, e.g., fog computing and edge computing. In such more decentralized computing paradigms, the location and resource for code execution and data storage of end applications could also be optionally distributed among different places or machines. In this paper, we position that this situation requires a new transparent and usercentric approach to unify the resource management and code scheduling from the perspective of end users. We elaborate our vision and propose a software-defined code scheduling framework. The proposed framework allows the code execution or data storage of end applications to be adaptively done at appropriate machines under the help of a performance and capacity monitoring facility, intelligently improving application performance for end users. A pilot system and preliminary results show the advantage of the framework and thus the advocated vision for end users.Recently, several novel computing paradigms are proposed, e.g., fog computing and edge computing. In such more decentralized computing paradigms, the location and resource for code execution and data storage of end applications could also be optionally distributed among different places or machines. In this paper, we position that this situation requires a new transparent and usercentric approach to unify the resource management and code scheduling from the perspective of end users. We elaborate our vision and propose a software-defined code scheduling framework. The proposed framework allows the code execution or data storage of end applications to be adaptively done at appropriate machines under the help of a performance and capacity monitoring facility, intelligently improving application performance for end users. A pilot system and preliminary results show the advantage of the framework and thus the advocated vision for end users.

关 键 词：cloud computing fog computing edge computing mobile edge computing resource management and code scheduling 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]