外文摘要：In this study, we propose a novel cloud-edge collaborative task assignment model for smart farms that consists of a cloud server, m edge servers, and n sensors. The edge servers rely solely on solar-generated energy, which is limited, whereas the cloud server has access to a limitless amount of energy supplied by the smart grid. Each entire task from a sensor is processed by either an edge server or the cloud server. We consider the task to be unsplittable. Building on the algorithm for the multimachine job scheduling problem, we develop a corresponding approximation algorithm. In addition, we propose a new discrete heuristic based on the dwarf mongoose optimization algorithmm, named the discrete dwarf mongoose optimization algorithm, and we utilize the proposed approximation algorithm to improve the convergence speed of this heuristic while yielding better solutions. In this study, we consider task sets with heavy tasks independently, where a heavy task is a task that requires many computing resources to process. If such tasks are assigned as ordinary tasks, the assignment results may be poor. Therefore, we propose a new method to solve this kind of problem.
外文关键词：Cloud-edge collaborative;Unsplittable;Approximation algorithm;Dwarf mongoose algorithm;Heavy tasks
作者：Xu, Minzhi;Li, Weidong;Zhang, Xuejie;Su, Qian
作者单位：Yunnan Univ
期刊名称：CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
期刊影响因子：0.0
出版年份：2024
出版刊次：27(5)
原文传递申请：江苏省科技资源（工程技术文献）统筹服务平台