Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (2): 503-508.DOI: 10.11772/j.issn.1001-9081.2017081971

Previous Articles     Next Articles

Service function chain construction method based on node utility maximization

ZHANG Chuanhao1,2, ZHOU Qiao2   

  1. 1. Police Technology Department, Railway Police College, Zhengzhou Henan 450053, China;
    2. China National Digital Switching System Engineering & Technological R & D Center, Zhengzhou Henan 450002, China
  • Received:2017-08-15 Revised:2017-10-13 Online:2018-02-10 Published:2018-02-10
  • Supported by:
    This work is partially supported by the National Basic Research Program of China (2012CB315901), the Technology Research Project of Chinese Ministry of Public Security (2016JSYJB38), the Foundation of Henan Science and Technology Department (172102210441).

节点效用最大化的服务功能链构建方法

张传浩1,2, 周桥2   

  1. 1. 铁道警察学院 公安技术系, 郑州 450053;
    2. 国家数字交换系统工程技术研究中心, 郑州 450002
  • 通讯作者: 张传浩
  • 作者简介:张传浩(1979-),男,河南郑州人,讲师,博士,CCF会员,主要研究方向:网络信息安全、网络功能虚拟化;周桥(1993-),男,湖北钟祥人,硕士,主要研究方向:网络功能虚拟化、服务功能链。
  • 基金资助:
    国家973计划项目(2012CB315901);公安部技术研究计划项目(2016JSYJB38);河南省科技厅基金项目(172102210441)。

Abstract: Networks heavily rely on middlebox to provide critical service functions, with the development of Software Defined Network (SDN) and Network Function Virtualization (NFV) technology, how to use the new technology to deploy middleboxs and guide flow through a specific sequence of middleboxs to complete the service function chain is still a problem to be solved. A construction method based on the optimal node utility maximization, namely NUM (Node Utility Maximization), was proposed for the service function chain construction problem, which took into account the deployment and steering of the virtual middleboxs in the meantime. Firstly, a service function chain collaborative construction mechanism was designed based on SDN+NFV technology. Secondly, the node selection model and the utility maximization model were introduced in this mechanism, according to the solution of middleware box deployment and traffic guidance problem. Finally, the model was solved by applying Tabu search-combined simulated annealing algorithm. The simulation results show that the proposed method NUM is superior to the traditional algorithm in terms of construction time, success rate and network congestion rate, and the utility of the nodes is improved by about 20% by using the proposed service function chain construction method.

Key words: Software Defined Network (SDN), Network Function Virtualization (NFV), middlebox, service function chain construction, Simulated Annealing (SA)

摘要: 网络严重依赖中间件盒子来提供关键的服务功能,随着软件定义网络(SDN)与网络功能虚拟化(NFV)技术的发展,如何利用新技术部署中间件盒子并引导流量通过特定顺序的中间件盒子完成服务功能链构建成为亟待解决的问题。针对SDN+NFV环境下的服务功能链构建问题,提出一种节点效用最大化的服务功能链协同构建方法NUM(Node Utility Maximization)。首先结合SDN+NFV技术设计了一种服务功能链协同构建机制;其次,根据机制中资源处理器所解决的中间件盒子部署和流量引导问题,建立了节点选择模型和效用最大化模型;最后,利用禁忌搜索改进组合模拟退火算法对该模型进行求解。仿真实验表明,NUM方法与主流方法相比在构建时间、构建成功率以及网络拥塞率上具有优越性,同时采用所提出的服务链构建方法在节点效用上提高约20%。

关键词: 软件定义网络, 网络功能虚拟化, 中间件盒子, 服务功能链构建, 模拟退火

CLC Number: