An Extensible Network Monitor Library Based On Multiple Thread

Nowadays,the demand for network communication is growing rapidly.An in-depth understanding of the Internet traffic profile is indispensable.Many network monitoring applications claim the connection-oriented analysis.However,many existing capture tools are too low-performance to cope with the large real-time network traffic flow.Though some tools using traffic sampling can handle heavy network stream,there is information loss.While some parallel methods have been adopted to optimize the performance,there are insufficient extendible easy-use libraries.To feed this demand,this paper proposes a flexible parallel network library(Noff)which has the ability to process the network traffic at a speed of up to 6Gbps.Noff supports two kinds of parallel processing procedures on the multi-core architecture to improve system compatibility and performance.Furthermore,Noff proposes a timing wheel scheme to process some network abnormity.For extendibility,Noff also offers many flexible callback interfaces.To arrive a better optimization effect,some open source optimization software methods are utilized to make a detail optimization on memory allocation and thread dispatch.Some well-designed experiments have been conducted to validate the performance of Noff.Then some detail comparisons have been done among Noff and many previous methods to analyze the merits and demerits.Meanwhile,this paper verifies the optimization effect of multiple thread,tc Malloc and many other methods.Finally,this paper makes a conclusion on the value of Noff and proposes some ideas of improvement.