Network flows' information statistics based on TCBF algorithm

Aiming at the problems of traditional network traffic record algorithm is easy to overflow and update frequently,this paper proposed a new algorithm based on time bloom filter counting bloom filter( TCBF) to record hight speed network traffic information in time. On the one hand,time bloom filter sampled the mice flows by using of the timeout characteristics,on the other hand,counting bloom filter filtrated the packets of large flows by using of network traffic distribution showed heavytailed characteristics. It analyzed the complexity and false positive rate of the algorithm. It analyzed the effect of flow information statistical accuracy and sampling rate for parameter configuration through simulation. The theoretical analysis and the simulation result indicate that compare to bloom filter,the TCBF agorithm can generate the statistics of the network flows immediately and accurately under the condition of using less storage space,and satisfies the need of actual measurement.