摘要
针对云计算虚拟化资源中,提高资源利用率、负载均衡度的问题,在蚁群算法的基础上,提出云中节点间负载均衡的改进算法。前向蚂蚁检测节点的类型、记录节点信息,遇到负载节点时留下觅食信息素;后向蚂蚁依据循迹信息素追溯回负载节点,合理分配超载节点任务。所有蚂蚁不再更新自己的结果集,而是致力于更新单个结果集,在搜索过程中依据节点类型动态地修改路径信息素。在Cloudsim平台下进行的仿真实验验证了改进算法的有效性。
Aiming at improving the utilization of virtualization resources and solving the load balancing problem in cloud computing,a load balancing algorithm within nodes in cloud network based on the ant colony algorithm was proposed.The forward ant detected node types,recorded node information and left foraging pheromone while meeting the overload node.Based on tracking pheromone,the backward ant backed to the loading node and allocated the overload node’s tasks rationally.All the ants don’t update their result sets,but make efforts to update a single result set,at the same time change the path pheromone based on the node type dynamically in the searching process.Simulation on the Cloudsim platform verifies the validity of the improved algorithm.
出处
《计算机工程与设计》
CSCD
北大核心
2014年第12期4095-4099,共5页
Computer Engineering and Design
基金
山西省科技攻关基金项目(20120321024-02)
关键词
蚁群算法
云计算
负载均衡
信息素
云仿真
ant colony optimization
cloud computing
load balancing
pheromone
cloud simulation