高效节能虚拟网的节点链路选择标准与映射算法研究

发布时间：2018-04-02 07:17

本文选题：网络虚拟化　切入点：能耗感知的虚拟网映射　出处：《郑州大学》2016年博士论文

【摘要】：随着互联网规模的增长,超额资源供给和冗余的系统设计原则带来的能源过度使用问题愈加严重,而网络虚拟化为网络的创新和可持续发展提供了有力的支持,同时也使得考虑节能因素的网络部署成为可能。网络虚拟化的核心思想是在同一个物理网络基础设施之上,为不同的业务构建各自的虚拟网,其中的关键问题是如何将虚拟网映射至物理网。虚拟网映射首先要满足不同的业务需求,还要根据网络的动态变化优化网络资源的配置,减少底层网络的能量消耗。节能目标的虚拟网络部署是实现网络虚拟化的重要目标之一,在节点和链路的节能选择标准以及节能映射算法的组织结构上,存在改进空间。本文工作如下:1.提出了一种纳入时间因素的虚拟网节能映射(TEAVNE)算法。该算法在选择虚拟节点或链路的映射目标时,最大化地减少了物理节点或链路的使用时长。仿真结果说明,与已有算法相比,该算法达到了更有效的节能目的;在资源充足的场景下,效果更为显著。2.在TEAVNE算法的基础上,提出了一种限制集中映射的虚拟网节能映射(EEVNE-CCM)算法。该算法在选择虚拟节点或链路的映射目标时,通过最小化最大资源利用率的方式提前开启部分节点和链路,以达到实现高效节能的目的。仿真结果表明,与TEAVNE算法相比,该算法在优化网络资源使用的同时,获得了更加明显的节能效果。3.提出了一种基于自适应协同进化粒子群的虚拟网节能映射算法(ACCPSO)。该算法针对具有离散解空间的虚拟网节能映射问题,通过对粒子位置聚集程度的分析,定义了聚集度,用于搜索最优解。仿真结果表明,与原始的粒子群算法相比,在相同的搜索次数下,该算法能够收敛到更优的节能映射方案。
[Abstract]:With the growth of the scale of the Internet, the over-use of energy brought by the excess supply of resources and redundant system design principles is becoming more and more serious, and the virtual network becomes the innovation and sustainable development of the network to provide strong support. The core idea of network virtualization is to build their own virtual networks for different services on the same physical network infrastructure. The key problem is how to map the virtual network to the physical network. The virtual network deployment of energy saving target is one of the important goals to realize network virtualization. In the aspect of energy saving selection standard of node and link and the organization structure of energy saving mapping algorithm, There is room for improvement. The work of this paper is as follows: 1. This paper proposes an energy saving mapping algorithm for virtual networks, which takes into account the time factor. The algorithm selects the mapping targets of virtual nodes or links. The simulation results show that compared with the existing algorithms, the algorithm achieves more effective energy saving purpose, and the effect is more significant in the scenario with sufficient resources. 2. Based on the TEAVNE algorithm, the simulation results show that the proposed algorithm is more effective than the existing algorithms. In this paper, an energy saving mapping EEVNE-CCM algorithm for virtual networks is proposed, in which some nodes and links are opened in advance by minimizing the maximum resource utilization when selecting the mapping targets of virtual nodes or links. The simulation results show that compared with the TEAVNE algorithm, the algorithm not only optimizes the use of network resources, but also optimizes the use of network resources. A new energy saving mapping algorithm based on adaptive coevolutionary particle swarm optimization (APCS) is proposed. The algorithm is aimed at the energy saving mapping problem of virtual networks with discrete solution space. By analyzing the aggregation degree of particle position, the aggregation degree is defined and used to search the optimal solution. The simulation results show that compared with the original particle swarm optimization algorithm, the algorithm can converge to a better energy saving mapping scheme under the same search times.
【学位授予单位】：郑州大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP393.01

【相似文献】