摘要
针对智能工厂中的边缘算力资源调度问题,提出基于边缘计算的多目标任务调度模型;以最小化任务时延和任务能耗为目标函数,添加传输时延、计算时延、排队时延、迁移时延、传输能耗、计算能耗、迁移能耗等7项约束条件建模;为了解决多目标调度求解问题,提出基于任务紧急度的资源调度与分配算法,并设置迁移判断中服务器选择和任务排序方法;以占地面积为1.2×104 m 2的智能工厂的边云设备为实验环境,对比基于任务紧急度的资源调度与分配算法与随机接入资源分配算法、迭代资源分配算法的平均任务完成率。结果表明,基于任务紧急度的资源调度与分配算法的服务器任务完成率较其他2种算法分别提高48、5个百分点,大多数服务器的任务丢弃率被降至2%~4%,并且对系统负载均衡等问题的处理效果显著。
Aiming at the scheduling problem of edge computing resources in industrial intelligent factories,a multi-objective task scheduling model based on edge computing was proposed.With the objective function of minimizing task delay and task energy consumption,seven constraints such as transmission delay,computing delay,queuing delay,migration delay,transmission energy consumption,computing energy consumption and migration energy consumption were added for modeling.In order to solve the multi-objective scheduling problem,a resource scheduling and allocation algorithm based on task urgency(RSA_TU)was proposed,and the methods of server selection and task sequencing in migration judgment were set.Taking the edge cloud equipment of a smart factory covering an area of 12000 m 2 as an experimental environment,RSA_TU algorithm was compared with the random-access resource allocation algorithm and iterative resource allocation algorithm based on greedy search in average mandate completion rate.The results show that the server task completion rate of RSA_TU algorithm is improved by 48 percentage points and 5 percentage points respectively compared with the other two algorithms,and the discard rate of most servers is reduced to 2%to 4%,which reflects the remarkable effect of the proposed algorithm on system load balancing and other issues.
作者
谢燕瑜
陈兰荪
XIE Yanyu;CHEN Lansun(School of General Education,Minnan Science and Technology College,Quanzhou 362332,Fujian,China;Academy of Mathematics and Systems Science,Chinese Academy of Sciences,Beijing 100190,China)
出处
《济南大学学报(自然科学版)》
北大核心
2025年第1期117-122,共6页
Journal of University of Jinan(Science and Technology)
基金
国家自然科学基金项目(12071407)。
关键词
生产任务调度
多目标优化
边缘计算
边云协同
production task scheduling
multi-objective optimization
edge computing
edge-cloud collaboration
