基于条件反亲和度的网络功能虚拟化资源分配算法

发布时间：2018-04-10 22:05

本文选题：条件反亲和度 + 网络功能虚拟化　；参考：《信息与控制》2017年05期

【摘要】：为了提高网络功能虚拟化(network function virtualization,NFV)资源分配的成功率,针对运营商网络的高可靠性要求,提出基于条件反亲和度的虚拟网络功能映射(virtual network function embedding,VNFE)算法,实现NFV资源分配.首先定义条件反亲和度,放宽NFV资源分配技术反亲和性设计的约束条件;其次,建立双备份VNFE算法模型,描述运营商网络NFV资源分配技术问题;最后,使用粒子群优化算法获得VNFE的解,解决NFV资源分配问题.仿真结果表明,所提算法能够满足用户对网络可靠性的要求,具有较高的资源利用,并将NFV资源分配的成功率提高约46%.
[Abstract]:In order to improve the success rate of network function virtualization (function) resource allocation in network virtualization, aiming at the high reliability requirement of operator network, a virtual network function embedding VNFE (virtual network function embedding) algorithm based on conditional anti-affinity is proposed to implement NFV resource allocation.Firstly, we define the degree of conditional anti-affinity and relax the constraints of anti-affinity design of NFV resource allocation technology. Secondly, we establish a double-backup VNFE algorithm model to describe the technical problems of NFV resource allocation in operator network.Particle swarm optimization (PSO) algorithm is used to obtain the solution of VNFE and to solve the problem of NFV resource allocation.Simulation results show that the proposed algorithm can meet the requirements of users for network reliability, has a higher resource utilization, and increases the success rate of NFV resource allocation by about 46.1%.
【作者单位】：西安培华学院中兴电信学院;西安交通大学电信学院;
【基金】：陕西省高等教育教学改革研究资助项目(15BY122)
【分类号】：TP393.0

【相似文献】