云平台网络负载均衡调度仿真研究

发布时间：2018-11-11 09:12

【摘要】：对云平台网络负载均衡调度,可以提高云平台网络资源的管理能力。进行负载均衡调度时,应利用云平台资源负载均衡模型计算调度所需的时间、费用、安全函数等参数,但传统方法依据云平台负载变化情况调整后端服务器集群的处理能力,进行数据多线程负载均衡调度,但不能建立云平台资源负载均衡模型,无法计算调度所需要的各种参数,耗时较长且调度性能差。提出云平台网络数据多线程负载均衡调度方法,首先建立云平台环境和异构集群并行计算熵矩阵,定义迁移虚拟机、迁移目标物理节点、源物理节点判断方式,并组建云平台资源负载均衡模型,获取云平台环境下的任务负载所需要的时间、可靠性、费用、安全性函数,将云平台中的数据资源负载均衡调度问题转化为离线空间的优化问题,结合人工萤火虫算法和混沌算法对该问题进行求解,利用其结果完成对云平台网络数据负载均衡调度。仿真结果表明,所提算法均衡性较强,可快速有效实现云平台网络数据多线程负载均衡调度。
[Abstract]:The network load balancing scheduling of cloud platform can improve the management ability of cloud platform network resource. In load balancing scheduling, the resource load balancing model of cloud platform should be used to calculate the time, cost, security function and other parameters of scheduling, but the traditional method adjusts the processing ability of back-end server cluster according to the load change of cloud platform. Multithreaded data load balancing scheduling, but can not establish a cloud platform resource load balancing model, can not calculate the various parameters required for scheduling, time-consuming and poor scheduling performance. A multi-thread load balancing scheduling method for cloud platform network data is proposed. Firstly, the cloud platform environment and heterogeneous cluster parallel computing entropy matrix are established, and the migration virtual machine, the target physical node and the source physical node are defined. The resource load balancing model of cloud platform is established to obtain the time, reliability, cost and security function of the task load in the cloud platform environment. The load balancing scheduling problem of data resources in cloud platform is transformed into an optimization problem in off-line space. The problem is solved by combining artificial firefly algorithm and chaos algorithm, and the results are used to complete the data load balancing scheduling of cloud platform network. The simulation results show that the proposed algorithm has strong equalization and can quickly and effectively realize multi-thread load balancing scheduling of cloud network data.
【作者单位】：内蒙古民族大学数学学院;
【分类号】：TP391.9;TP393.0

【相似文献】