摘要
在对云环境下数据流量异常行为网络负载均衡优化设计中,由于云环境下的数据流量异常行为具有随机性和突发性,造成数据流量异常的端口出现无规律性。传统的方法在进行分析的过程中,受到无规律性的影响不能对异常流量数据进行准确的分类,导致负载不均、效率低,时间长的问题。提出采用并行计算熵的任务负载均衡方法,将服务请求划分至内部处理负荷最轻的服务器。分析了云计算环境下任务负载均衡模型,依据样本均值与标准差对流量特性的总体均值建立一个置信区间,判断流量异常行为进行。计算各节点性能的量化值,求出节点相对负载和系统相应的并行计算熵。判断系统并行计算熵与阈值之间的关系,对各节点进行迁移和更新,直至满足阈值关系。仿真结果表明,所提方法可有效实现任务划分,使云计算环境下资源节点的均衡使用,同时减少任务完成时间。
A task load balancing method based on parallel computing entropy was proposed in the paper,and the service request is partitioned to the server with lightest internal processing load. The task load balancing model in cloud computing environment was analyzed. Based on the mean and standard deviation of samples,a confidence interval was established for the population mean of the traffic characteristics to determine the traffic abnormality behavior. The quantization value of each node＇s performance was calculated,and the relative load of nodes and the corresponding parallel computing entropy of system were obtained. The relationship between the parallel computing entropy and threshold value of system was judged,and the nodes were migrated and updated to meet the threshold value relationship. Simulation results show that the proposed method can effectively achieve the task partition,so that the resource nodes in the cloud computing environment are used evenly,and the task completion time is reduced.
出处
《计算机仿真》
CSCD
北大核心
2016年第2期345-349,共5页
Computer Simulation
基金
河南省高等学校重点科研项目(15B520036)
关键词
云计算
流量异常
任务负载均衡
Cloud computing
Traffic anomaly
Task load balancing
