一种面向SDN/NFV的多目标优化服务功能链部署策略

A multi-objective optimization service function chain deployment strategy oriented SDN/NFV

针对当前基于SDN/NFV的服务功能链(SFC)部署策略无法在优化网络性能的同时降低资源消耗的问题,提出一种基于模拟退火粒子群优化(PSO)算法的多目标优化部署策略。首先,综合考虑网络节点和逻辑链路因素,建立以时延、平均路径长度和负载强度为优化目标的多目标线性规划模型;其次,采用粒子群算法求解服务链部署问题,从而构建出一条近似最优的服务路径。仿真实验结果表明:所提出的策略在降低负载强度、映射代价和总时延的同时,提高了服务请求接受率。

Aiming at the problem of deployment strategy of service function chain(SFC)based on SDN/NFV cannot optimize network performance,at the same time,reduce resource consumption,a global optimal deployment strategy based on simulated annealing particle swarm optimization(PSO)algorithm is proposed.Firstly,consider the factors of network node and logical link synthetically to build a multi-objective linear programming model with time delay,average path length and load strength as optimization goals.After that,the particle swarm algorithm is used to solve the service chain deployment problem,thereby construct an approximately optimal service path.Simulation experiments show that the proposed strategy reduces the load intensity,mapping cost,and total delay,at the same time,improve the service request acceptance rate.