云计算环境下基于改进GSO的目标主机选择算法

A Selection Algorithm of Destination Host Based on Improved Glowworm Swarm Optimization Algorithm in Cloud Computing Environment

目标主机的选择是虚拟机动态迁移过程中的重要阶段,是实现负载均衡的关键。针对基本萤火虫算法存在的精度不高、收敛较慢的问题,提出了一种改进的萤火虫优化算法,用于解决虚拟机迁移时虚拟机和目标物理主机的映射问题,实现多目标最优求解。该算法通过引入步长调整因子,能够动态调整移动步长,克服了步长过大或过小导致的精度不高、后期收敛较慢的缺点。全面考虑物理主机负载指标,建立负载均衡模型,将萤火虫算法中个体与节点资源相对应,利用萤火虫发光机制寻优求解,以实现目标主机的优化选择。仿真实验表明,该算法能够快速完成目标主机的选择,有效平衡系统资源,实现数据中心负载均衡。

The selection of destination host is an important stage in the dynamic migration of virtual machines,and is the key to realize load balancing. In order to overcome the shortcomings of basic glowworm swarm optimization algorithm including low accuracy and slow convergence speed,an improved glowworm swarm optimization algorithm is proposed to solve the mapping problem between the virtual machine and the target physical host when the virtual machine is migrated,and the multiobjective optimal solution is realized. By introducing the step adjustment factor,the algorithm can dynamically adjust the moving step and overcome the shortcomings of low accuracy and slow convergence speed caused by too large step or too small step. Considering the physical load index,the load balancing model is established,and the individual and node resources in the improved glowworm swarm optimization algorithm are matched with each other,and the optimal selection of the destination host is realized by using the luminous mechanism of fireflies. The simulation experiment shows that the improved algorithm can select the destination host quickly,balance the system resource effectively and realize the load balancing of data center.