基于粒子群和改进蚁群算法的云计算任务调度
Cloud computing task scheduling based on particle swarm algorithm and improved ant colony algorithm
任小强 1聂清彬 1王浩宇 1林慧琼1
作者信息
- 1. 西南交通大学希望学院信息工程系,四川成都 610400
- 折叠
摘要
针对目前云计算任务调度方法的效率较低和日益多样化的用户服务质量需求等问题,提出一种将粒子群算法和改进蚁群算法结合的混合粒子群蚁群算法(HPSO-ACO),包括建立云计算任务调度模型、用户服务质量模型及虚拟资源节点模型.利用离散型粒子群算法,得到初始解集,转化为蚁群算法信息素的初始值,通过改进蚁群算法的寻径规则和信息素更新规则,得到最终解.通过仿真实验将粒子群算法、蚁群算法和HPSO-ACO算法进行比较,其结果表明,HPSO-ACO算法有效且可行,能够减少任务完成时间和降低完成成本,满足用户服务质量要求.
Abstract
Aiming at the low efficiency of the current cloud computing task scheduling method and the increasingly diverse user quality of service requirements,the particle swarm-ant colony algorithm that combined the particle swarm algorithm and the im-proved ant colony algorithm to perform task scheduling was proposed,including cloud computing task scheduling model,user quality of service model and virtual machine resource model.The discrete particle swarm algorithm was used to obtain the initial solution,the results of particle swarm algorithm were converted into the value of ant colony algorithm's initial pheromone.By improving the path finding rule and pheromone update rule of the ant colony algorithm,the final solution of the cloud computing task scheduling problem was obtained.The particle swarm algorithm,ant colony algorithm and the proposed algorithm were compared by simulation experiments.Experimental results show that the feasibility and validity of the proposed algorithm is veri-fied.It can reduce the task completion time and satisfy user quality of service.
关键词
云计算/有向无环图/用户服务质量/蚁群算法/信息素/粒子群算法/任务调度方案Key words
cloud computing/directed acyclic graph/quality of service/ant colony algorithm/pheromone/particle swarm algo-rithm/task scheduling scheme引用本文复制引用
基金项目
教育部产学合作协同育人教学内容和课程体系改革基金(202102041003)
成都市交通+旅游大数据应用技术研究基金(2022117)
西南交通大学希望学院一流本科课程建设项目(2112056)
出版年
2024
NETL
NSTL
万方数据