Research Area:  Edge Computing
The workflow, in edge computing, is capable of orderly managing a set of computing tasks, via setting up a structured calculation process and flow. However in running time, the resources at edge nodes are often released by other workflows, which results in additional uncertainties on top of those caused by performance degradation and service failure. In addressing these new uncertainties, we propose in this paper an uncertainty-aware resource provisioning (UARP) method for scheduling workflow in the software-defined network-based(SDN-based) edge computing environment. The proposed UARP ensures optimal in that the non-dominated sorting genetic algorithm-III (NSGA-III) is employed to elaborate the workflow scheduling strategy. Extensive experiments are carried out for method evaluations and the results show that the UARP method is effective in reducing uncertainty, shortening processing time and bringing down energy consumption.
Keywords:  
Author(s) Name:  Hao Cao; Xiaolong Xu; Qingxiang Liu; Yuan Xue; Lianyong Qi
Journal name:  
Conferrence name:  IEEE International Conference On Trust, Security And Privacy In Computing And Communications/13th IEEE International Conference On Big Data Science And Engineering
Publisher name:  IEEE
DOI:  10.1109/TrustCom/BigDataSE.2019.00105
Volume Information:  
Paper Link:   https://ieeexplore.ieee.org/document/8887332