Title :
VM-based Architecture for Network Monitoring and Analysis
Author :
Li, Qiang ; Hao, Qinfen ; Xiao, Limin ; Li, Zhoujun
Author_Institution :
Sch. of Comput. Sci., Beihang Univ., Beijing
Abstract :
A single physical machine provides multiple network monitoring and analysis services (e.g., IDS, QoS) which are installed on the same operating system. Isolation between services is weak and it is difficult to decide the optimum allocation of resources for each service. This paper presents a virtual-machine-based architecture for network traffic monitoring and analysis. Through virtualization, a machine under the architecture logically is divided into one host, one virtual machine monitor (VMM) and multiple virtual machines. The host is responsible for capturing network traffic, and multiplexing it to multiple virtual machines. Each virtual machine hosts a service. VMM performs functions such as isolating services and resolving the conflict between services. Compared with Xen, KVM is chosen as a VMM to implement the architecture. Some network optimizations of the architecture are given. Our evaluation results show that these optimizations can multiplex network traffic received by the host to all services, and improve the data receive performance of services by 67% compared to the architecture in which the traffic is directly transferred to virtual machines, instead of the host, and optimized methods are not adopted.
Keywords :
computer networks; telecommunication traffic; virtual machines; network optimization; network traffic analysis; network traffic monitoring; virtual machine monitor; virtual-machine-based architecture; virtualization; Computer architecture; Computer networks; Computer science; Computerized monitoring; Condition monitoring; Telecommunication traffic; Virtual machine monitors; Virtual machining; Virtual manufacturing; Voice mail; KVM; Virtual machine; Xen; network monitoring; virtualization;
Conference_Titel :
Young Computer Scientists, 2008. ICYCS 2008. The 9th International Conference for
Conference_Location :
Hunan
Print_ISBN :
978-0-7695-3398-8
Electronic_ISBN :
978-0-7695-3398-8
DOI :
10.1109/ICYCS.2008.445