面向虚实互联网络的链路采集技术研究

Research of Link Acquisition Technology for Virtual-reality Network

网络仿真是网络空间靶场的重要基础,面向仿真网络的数据采集与评估是关键环节。基于OpenStack复现形成的仿真网络,构建面向多样化仿真链路的分布式链路数据采集系统。研究面向宿主机内、间的虚拟链路和虚实互联等3类链路的协同采集技术、高效链路数据采集技术以及基于链路数据的效果评估技术。实验表明:所提方法在采集速率上相对于传统方法提升了88.48%,在CPU占用率上降低了45.29%;面向典型的LDDoS场景,可采集所需要的链路数据,并进行LDDoS的效果评估。

Network simulation is an important foundation for the cyberspace shooting range,where the data acquisition and evaluation for the emulation network is its key link.Based on the emulation network formed by OpenStack,a distributed link data acquisition system for the diversified emulation links is constructed.The paper studies the collaborative acquisition technology and the efficient link data acquisition technology of 3 kinds of links,such as the virtual link in the host,the virtual link between the host and the virtual-reality link,and the evaluation technology based on the link data.Experiments show that the proposed method improves the acquisition rate by 88.48%compared with the traditional method,and reduces the CPU usage by 45.29%.For a typical LDDoS scenario,the proposed method can collect the required link data and evaluate the effect of LDDoS.