Long stream recognition method, data flow measuring method and device thereof

Abstract

The embodiment of the invention relates to an identification method of long flows and a measuring method of data flow and a device thereof, wherein the identification method of long flows comprises assuring term number of business flow to be 1, wherein each term is corresponding to a prior counting type Blum filter, doing hash calculation for each term, generating hash address of each prior counting type Blum filter which is corresponding to each term, searching for a counter which is corresponding to each hash address which is corresponding to each term, circulating a counter which is corresponding to each prior counting type Blum filter, and judging the business flow is one long flow if the value of each counter in each term is bigger than and is equal to a preset threshold value or circulating time of a counter corresponding to each term is equal to the number of the prior counting type Blum filter, and returning to the step of confirming the term number of business flow to be 1 after adding 1 to the value of each counter in each term. The embodiment of the invention achieves the identification and statistic to long flows which are on line in real time in high-speed network, and is suitable for flow measuring requirements in high-speed network.

Description

technical field [0001] The invention relates to the technical field of computer communication, in particular to a long flow identification method, a data flow measurement method and equipment thereof. Background technique [0002] With the expansion of the scale of the modern Internet, the prominence of heterogeneity, and the diversity of network applications, people urgently need to understand and master its behavioral characteristics in a local and overall range in order to better understand, design and Manage the Internet. Network traffic measurement is the basis of traffic engineering and network behavior studies. It can measure the basic parameters of network traffic, such as packet length distribution, throughput, self-similarity coefficient, etc., and model and analyze the network according to the measurement results. In order to better grasp the basic characteristics of network behavior and ensure the healthy operation of the network. It is because of the above rea...

AI Technical Summary

Problems solved by technology

[0014] During the research and practice of the prior art, the inventors of the present invention found that in the existing implementation, the traffic measurement in the high-speed network faces the challenge of scalability, although CBF can support the insertion and hashing of set elements Query and statistical counting of collection elements, however, due to the low space utilization of CBF, it cannot identify and count real-time online long flows in high-speed networks, and cannot adapt to current and future flow measurement needs

Images