| With the rapid development of network information technology, the types of Internet services are increasing, and the network structures are increasing complicated . How to evaluate network quality, and how to position and solve the network fault quickly and efficiently are of vital importance. Therefore, active network performance measurement is becoming a research hotspot in the field of scientific research.Distributed measurement system is composed of probe and the control center. End users are able to assign the network quality monitoring task via webpage to monitor the network quality. The existing measurement systems mainly use probes located at local area network(LAN) , in which the number of nodes are limited and the structure of topology is simple. Moreover, the measure tasks are liminted in the network layer and application layer. However, the existing system will meet many difficulties such as node deployment and simplicity of measurement metrics while deployed in complex large-scale network.Therefore, the existing system optimization mechanism research has important significance.This paper proposed a node deployment algorithm based on detection strategy, which optimized the traffic load generated by active measurement. This paper also realized the cross layer multi-protocol correlation measurement function, which combined fault analysis with network measurement. Finally, this paper optimized the process of network measurement. The detection strategy proposed in this paper is based on the NS3 simulation network environment. The cross-layer multi-protocol correlation measurement is implemented on the Gperf measurement platform, and the experiment is realized in the real network environment. The results show that the method proposed in this paper can reduce the extra flow caused by the active measurement of the probe. And the method can integrate network layer measurement with application layer measurement into one process which is convenient for fault analysis. |
PDF Full Text Request