摘要
基于云计算环境下资源利用率低的问题,将一种新提出的金豺优化算法应用于云计算资源调度策略。研究采用Cloudsim作为仿真实验平台,以减少任务总完成时间为优化目标。实验中以猎物位置模拟任务对虚拟机序号的选择,在一定的迭代次数后根据猎物位置得出每个任务对虚拟机序号的最终选择和最终任务总完成时间。改变金豺优化算法的迭代次数并进行实验,结果表明,在迭代次数达100次时,金豺优化算法在云计算资源调度模型上的效果达到最优。将调用金豺优化算法与应用贪心算法和遗传算法下的实验结果进行对比,结果表明,在任务数量大于1000时,金豺优化算法在云计算资源调度模型上的效果优于贪心算法和遗传算法,效率相较于遗传算法提升了约20%。
Based on the problem of low resource utilization in the cloud computing environment,a newly proposed Golden Jackal optimization algorithm is applied to the cloud computing resource scheduling strategy.The research uses Cloudsim as a simulation experiment platform,with the optimization goal of reducing the total task completion time.In the experiment,the selection of the virtual machine serial number for each task is simulated using the prey location.After a certain number of iterations,the final selection of the virtual machine serial number for each task and the total completion time of the final task are obtained based on the prey location.After changing the iteration number of the Golden Jackal optimization algorithm and conducting experiments,the result shows that when the iteration number reaches 100,the Golden Jackal optimization algorithm achieves the optimal effect on the cloud computing resource scheduling model.Comparing the experimental results of using the Golden Jackal optimization algorithm with the experimental results of using the greedy algorithm and genetic algorithm,the result shows that when the number of tasks is greater than 1000,the Golden Jackal optimization algorithm outperforms the greedy algorithm and genetic algorithm in the cloud computing resource scheduling model,with an efficiency improvement of about 20% compared to genetic algorithm.
作者
李伟彦
董宝良
王凯
廉兰平
LI Weiyan;DONG Baoliang;WANG Kai;LIAN Lanping(North China Institute of Computing Technology,Beijing 100083,China;China Electronics Technology Group Corporation,Beijing 100846,China)
出处
《电子设计工程》
2023年第15期41-45,共5页
Electronic Design Engineering
关键词
云计算
金豺优化算法
资源调度
总完成时间
cloud computing
Golden Jackal optimization algorithm
resource scheduling
total completion time
