云平台网络负载均衡调度仿真研究

Simulation Research on Load Balancing Scheduling of Cloud Platform Network

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。

This article proposes a method for schedule of multithreading load balancing of date in cloud platform network. Firstly, the author built environment of cloud platform and heterogeneity cluster and calculated entropy matrix, then defined transfer virtual machine, physical node of transfer target and way of judgment of physical node. The research also built a model for load balancing of cloud platform environment to acquire function of time, reliability, cost, and security required by mission load under platform environment and converted problem of balancing scheduling of data information load into optimization problem of off - line space. Further, the research solved the problem integrated with artificial firefly algorithm and chaos algorithm. Finally, simulation results show that the algorithm mentioned above has better equilibrium. It can achieve the multithreading load balancing scheduling of network data in cloud platform rapidly and effectively.