摘要
随着互联网逐步深入人类的生活,人们对互联网的服务质量提出了更高的要求,这给互联网带来了巨大的挑战:一是移动网络、空间网络等异构网络以及各种异构终端的接入,网络规模不断扩大,导致网络服务管理复杂;二是互联网新应用的出现,给网络服务提出了更高的要求.但是,由于网络和应用缺乏相互的感知,严重影响了网络质量的提高.为了解决网络与应用相互独立的问题,需要建立网络信息平面来沟通网络和应用以提高网络服务质量.随着SDN(Software-Defined Networking)等网络新技术的发展和大数据思想深入影响到互联网的架构和通信模式,在网络架构中融入大数据驱动的网络信息平面以及密集计算驱动的分析方法已成为重要的研究方向.在现有的网络中,利用测量数据构建网络服务平台并通过数据驱动的方式优化网络资源的管理效率和提升用户体验,也成为当前工业界和学术界关注的热点.文中首先分析了网络信息平面构建的必然性,然后提出了一种大数据驱动的网络信息平面架构,分析了信息平面研发中面临的三大难题,最后,对解决这些难题提出了一些想法和建议.
As the Internet gradually deepening human life, people put forward higher requests to the Internet quality of service, this brings great challenges to the Internet. First, heterogeneous network and terminals are accessed the Internet, which makes the network scale continually expand and complicates on the network service management. Second, the emergence of new application has put forward higher requirements to the network service. But the lack of mutual awareness between the Internet and application, this impacts on the network to improve the quality of services. In order to solve above question, it needs to build the network information plane to communicate the Internet and application. With the development of SDN （Software-Defined Net- working） and other new technologies and big data thought deeply influenced to the architecture of the Internet and communication mode, big data-driven network information plane integrated into the network architecture and intensive computation-driven analysis method become an important research direction. In the existing network, constructing the network service platform and through big data-driven approach to optimize the efficiency of network resource management and enhance the user experience by using the measured data also become the focus of academic and industrial attention. This paper first analyzes the necessity of the construction of network infor- mation plane, and then puts forward the big data-driven network information plane architecture and analyzes three major problems faced in the development of information plane. Finally, we put forward some ideas and suggestions to solve these problems.
出处
《计算机学报》
EI
CSCD
北大核心
2016年第1期126-139,共14页
Chinese Journal of Computers
基金
国家"九七三"重点基础研究发展规划项目基金(2012CB315801)
国家自然科学基金(61222213
61170290)
清华大学自主科研计划--"新一代多媒体网络体系结构与关键技术研究"资助
关键词
大数据
网络信息平面
软件定义网络
big data
network information plane
Software-Defined Networking （SDN）
