摘要
在底层网络节点异构的环境中,能耗优化的虚拟网络映射问题并不是最小化工作节点和链路数。该文针对此问题,构建底层网络节点和链路的负载能耗模型,并以能耗最优为目标,建立虚拟网络映射问题的数学模型,提出一种能耗感知虚拟网络映射算法。该算法在节点映射阶段以最小化能耗和协调链路映射为原则,将虚拟节点映射至综合资源能力最大的底层节点上,并采用改进的能耗感知k最短路径法进行链路映射。仿真结果表明,该算法显著减少虚拟网络映射的能耗,且底层网络节点异构性越大,能耗优势更为明显。
The energy optimized virtual network embedding problem in the substrate network with heterogeneous nodes is not to minimize the number of working nodes and links. The load-based energy consumption models of the node and link in the substrate network are built, a mathematical model of the virtual network embedding problem is modeled in order to reduce energy consumption, and an energy-aware virtual network embedding heuristic algorithm is proposed. Based on the principles of energy optimization and coordination with link mapping, the virtual node is mapped onto the substrate node with the highest comprehensive resource capacity in the node mapping phase, and the link mapping phase is based on the energy-aware k shortest path algorithm. Simulation results show that the proposed algorithm reduces the energy consumption significantly, and the heterogeneity of substrate network nodes is greater, reducing the energy consumption is more obvious.
出处
《电子与信息学报》
EI
CSCD
北大核心
2015年第8期2021-2027,共7页
Journal of Electronics & Information Technology
基金
国家自然科学基金(51075395)
国家863计划项目(2013AA040604)资助课题
关键词
网络虚拟化
虚拟网络映射
能耗模型
资源能力
Network virtualization
Virtual network embedding
Energy consumption model
Resource capacity