基于烟花算法的云作业调度策略  

Cloud Job Scheduling Strategy Based on Fireworks Algorithms

作　　者：李强[1] 刘晓峰 孔德瑾[1] 秦华伟 吴瑞芳 LI Qiang;LIU Xiaofeng;KONG Dejin;QIN Huawei;WU Ruifang(College of Big Data,Shanxi Finance and Taxation College,Taiyuan 030024,China;College of Computer Science and Technology(College of Data Science),Taiyuan University of Technology,Taiyuan 030600,China)

机构地区：[1]山西省财政税务专科学校大数据学院,山西太原030024 [2]太原理工大学计算机科学与技术学院(大数据学院),山西太原030600

出　　处：《湖南科技大学学报（自然科学版）》2023年第4期88-96,共9页Journal of Hunan University of Science And Technology：Natural Science Edition

基　　金：国家自然科学基金资助项目(61502330);山西省高等学校科技创新项目资助(2020L0743;2022-676);山西省软科学计划研究项目资助(2016041008-5)。

摘　　要：典型的云作业调度策略不能满足现有云系统工作效率的需求,为了进一步提升其性能,提出一种基于烟花算法的调度策略.首先,分析限制云作业调度的节点性能指标,并将这些节点性能指标作为调度器的决策因素;然后,通过数学问题建模,使用模拟植物生长的算法改进烟花算法中爆炸烟花的分布方式,使其按照植物的生长方式分布烟花;最后,将所提算法与4种典型的云作业调度算法进行试验对比,分析所提算法的性能.结果表明:与典型的云作业调度算法相比,所提算法可以更好地提升系统的性能.Typical cloud job scheduling algorithms can not meet the efficiency requirements of existing cloud systems,and the improvement of algorithms needs to be further implemented,a cloud job scheduling strategy based on fireworks algorithms is proposed.Firstly,the performance indicators of nodes that restrict cloud job scheduling are analyzed,and are used as decision-making factors of the scheduler.Secondly,through mathematical problem modeling,the plant growth simulation algorithm is used to improve the distribution of explosive fireworks in traditional fireworks algorithm,so that fireworks can be distributed according to the growth mode of plants.Finally,four typical cloud job scheduling algorithms are compared with the proposed algorithm by experiments.Experiments show that the proposed algorithm can better improve the performance of the system than the typical scheduling algorithm for cloud job.

关 键 词：调度算法 模拟植物生长算法 云计算 负载均衡 烟花算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]