Accurate traffic replay based on interactive sequence and timestamp

Hao Wu, Hongri Liu, Bailing Wang, Guodong Xin · 2017

Existing traffic replay methods are mainly aim to generate a large number of network traffic per unit time, which neglect the time's authenticity of replay traffic. In order to generate the network traffic which is exactly the same as the real traffic in the target network, including packet's numbers, payloads, interactive orders and time series, this paper proposes a traffic replay method based on packet's interactive sequence and timestamp. Firstly, zero-copy technology is used to capture network traffic in real network, Then the flows are processed by data processing operations such as de-duplicating, ip-address mapping and filtering. Finally, network traffic is generated accurately in the target network by the replay program. The experimental results show that the traffic which is generated by our methods is completely consistent with the original traffic at packet's numbers, payloads, interaction orders. In addition, the similarity of the interaction time is averaged at 99.9%, which confirms that this method is correct and feasible.

Read the paper · More papers on PaperTik