基于分布式禁忌人工蜂群算法的云视频直播优化算法  

OPTIMIZATION ALGORITHM FOR LIVE CLOUD VIDEO BASED ON THE DISTRIBUTED TABU ARTIFICIAL BEE COLONY ALGORITHM

作　　者：梁建胜[1] 谭思敏 Liang Jiansheng;Tan Simin(Dongguan Polytechnic,Dongguan 523808,Guangdong,China;Cloud Computing Center of the Chinese Academy of Sciences,Dongguan 523808,Guangdong,China)

机构地区：[1]东莞职业技术学院,广东东莞523808 [2]东莞中国科学院云计算产业技术创新与育成中心,广东东莞523808

出　　处：《计算机应用与软件》2019年第12期122-130,149,共10页Computer Applications and Software

基　　金：广东省科技计划项目(2016B050502001);东莞市2019年哲学社会科学规划课题(2019ZZ08)

摘　　要：为了平衡热门视频直播服务的用户体验质量和系统总体成本,设计一种基于分布式禁忌人工蜂群算法的云视频直播优化算法。将禁忌搜索技术引入人工蜂群算法,对人工蜂群算法的全局搜索阶段和局部开发阶段均进行增强处理;以云服务站点和视频显示格式的选择为控制量,以视点的受欢迎度和观看者的可用带宽为约束条件,以最大化用户体验质量和最小化系统总成本为优化目标,建立云视频直播问题的模型;设计分布式的禁忌人工蜂群算法求解云视频直播问题的次优解,计算最优的控制方案。实验结果表明,该算法有效地降低了系统的总体成本,并且优于其他同类型的策略。To balance the quality of experience of users and total system cost of popular live video service,we propose an optimization algorithm for live cloud video based on the distributed tabu artificial bee colony algorithm.We introduced the tabu search technique into artificial bee colony algorithm,and enhanced the global exploration phase and local exploitation phase of artificial bee colony algorithm.Taking the selections of cloud service sites and video display formats as control variables,the popularity of views and the available bandwidths of viewers as constraint conditions,and the maximum of the quality of user experience and the minimum of the total cost of the system as the optimization objectives,we established the live cloud video model.A distributed tabu artificial bee colony algorithm was designed to solve the sub-optimal solution of the live cloud video problem and calculate the optimal control scheme.Experimental results show that this algorithm effectively reduces the total cost of the system,and is superior to the other similar algorithms.

关 键 词：云视频 云计算 人工蜂群算法 禁忌搜索 视频直播 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]