CN102404174A - Method for detecting network bandwidth - Google Patents

Abstract

The present invention relates to a method for detecting network bandwidth, comprising: a. using a default bandwidth to download a predetermined amount of data from a speed test website to obtain an initial bandwidth; b. a load balancing module and a flow control module apply the initial bandwidth to control; c. obtaining the existing bandwidth and the current maximum bandwidth according to the data flow; d. calculating the average value of the existing bandwidth and the current maximum bandwidth, and judging whether to use the calculated average value as the initial bandwidth to reset the bandwidth according to the calculated bandwidth and the change amount of the current bandwidth; e. looping from step b at intervals set to perform real-time bandwidth detection. The method of the present invention solves the problem that it is inconvenient for customers to configure routers in an environment where the external network bandwidth changes dynamically, and by automatically detecting the network bandwidth, the detected bandwidth is dynamically applied to flow control and load balancing, which greatly improves the accuracy of detecting the bandwidth and improves the flow control and load balancing effects.

Description

Method of Network Bandwidth Detection

technical field

The invention relates to a method for detecting network bandwidth.

Background technique

In the network environment with router access, due to the existence of various lines such as ADSL and static optical fiber, when configuring the router, users often need to set the router according to the actual bandwidth to ensure flow control, load balancing, etc. The function works normally. However, usually some customers do not know the actual bandwidth, or there is a gap between the actual bandwidth and the nominal bandwidth, which often leads to problems in flow control and load balancing.

Contents of the invention

In view of the above problems, the present invention provides a method for network bandwidth detection, which automatically detects network bandwidth, and dynamically applies the detected bandwidth to flow control and load balancing, greatly improving the accuracy of bandwidth detection and improving Flow control and load balancing effects.

The method for network bandwidth detection of the present invention comprises:

a. Use the default bandwidth to download a predetermined amount of data from the speed measurement website, and obtain the initial bandwidth according to the quotient of the size of the data amount and the corresponding download time;

b. The load balancing module and the flow control module apply the initial bandwidth to the control to realize load balancing and flow control functions;

c. Record the current data flow within 1 second at least once, obtain the current average bandwidth and the current maximum bandwidth of each record according to the data flow, average the current bandwidth averages of each time to obtain the existing bandwidth, and obtain the current bandwidth for each time The average of the current maximum bandwidth to the current maximum bandwidth;

d. Calculate the average value of the existing bandwidth and the current maximum bandwidth average value, if the difference between the average value and the current bandwidth configured by the current system > the set threshold, then use the average value as the initial bandwidth bandwidth reset;

e. The time set at intervals is executed cyclically from step b for real-time detection of bandwidth.

When the router is configured, the initial external network bandwidth is calculated according to the actual data obtained by measuring the speed of the specified website. During the subsequent use of the router, the external network bandwidth is regularly sampled. According to the average bandwidth used and the continuous peak value, Continuously detect the test bandwidth, adjust the most realistic bandwidth value and configure the router in real time, thereby greatly improving the accuracy of bandwidth detection and improving the effect of flow control and load balancing.

Preferably, in step c, the current data traffic within 1 second is continuously recorded 5 times, and the current average bandwidth and the current maximum bandwidth of each record are calculated according to the data volume of the 5 times. From the perspective of probability, the more sampled values are, the more accurate the final value will be. However, taking into account the efficiency, it is preferable to perform 5 samples every time this step is performed.

Preferably, step c is repeated 5 times before entering step d.

Preferably, the existing bandwidth described in step c is calculated by the last 10 current bandwidth mean values; the current maximum bandwidth mean value described in step c is calculated by the last 10 current maximum bandwidth values to obtain a more accurate existing bandwidth and the current maximum bandwidth average.

Further, since there is a certain difference between the actual bandwidth peak value and the standard peak value, the maximum average bandwidth value described in step d is compensated according to the sample variance, so that the result is more in line with the actual bandwidth value.

Preferably, the set threshold in step d is 10% of the current bandwidth configured by the current system, and the set threshold can also be adjusted according to actual conditions.

Preferably, the setting time described in step e is 300 seconds, each step of the method of the present invention is repeated every 300 seconds, and the current external network bandwidth is detected in real time to realize dynamic configuration.

The method of the present invention solves the problem that it is inconvenient for customers to configure routers in an environment where the bandwidth of the external network changes dynamically, and through automatic detection of network bandwidth, the detected bandwidth is dynamically applied to flow control and load balancing, which is very convenient Greatly improve the accuracy of bandwidth detection, and improve the effect of flow control and load balancing.

The above content of the present invention will be further described in detail below in conjunction with the specific implementation manners of the embodiments shown in the accompanying drawings. However, this should not be construed as limiting the scope of the above-mentioned subject matter of the present invention to the following examples. Without departing from the above-mentioned technical idea of the present invention, various replacements or changes made according to common technical knowledge and customary means in this field shall be included in the scope of the present invention.

Description of drawings

Fig. 1 is a flow chart of the method for network bandwidth detection in the present invention.

Detailed ways

As shown in Figure 1, the method for network bandwidth detection of the present invention includes:

a. When the device is started and the initial network connection is completed, the intranet data connection is prohibited to avoid data interference. Use the default 10M (megabytes) bandwidth to download a predetermined amount of data from the speed measurement website, and obtain the initial bandwidth B ₀ according to the quotient of the size of the data amount and the corresponding download time T, that is, the initial bandwidth B ₀ (bit/s) = download Data volume M(bit)/download duration T(s), duration T=download end time T2-download start time T1;

b. The load balancing module and the flow control module apply the initial bandwidth _B0 to the control to realize load balancing and flow control functions;

c. Continuously record 5 times the current data traffic within 1 second, which are recorded as M1, M2, M3, M4, M5 respectively, and calculate the current average bandwidth B _n (bit/s)=(M1 +M2+M3+M4+M5)/5, and the current maximum bandwidth BM _n =Max(M1, M2, M3, M4, M5). After repeated execution 5 times, the current bandwidth is obtained by averaging the last 10 current bandwidth averages B _n , and then the current maximum bandwidth average is obtained by averaging the latest 10 current maximum bandwidth BM _n . Current maximum bandwidth average BM=Average{BMn- ₀ , BMn- ₁ , BMn- ₂ , BMn- ₃ , BMn- _{4 ,} BMn- ₅ , BMn- 6 , BMn- ₇ , BMn- ₈ , BMn- ₉ }. Every time it is recorded, the value of the subscript n is incremented by 1, and the calculated value is recorded as historical data;

d. Compensate the maximum bandwidth mean BM according to the sample variance: sample variance S=[(Bn- ₀ -X)^2+(Bn- ₁ -X)^2....+(Bn- ₉ -X)^ 2]/(10-1), X means sample mean=(Bn- ₀ +Bn-1+...+Bn- ₉ )/10. Calculate the average value of the existing bandwidth and the current maximum bandwidth mean value BM, if the difference between the average value and the bandwidth configured by the current system is > 10% of the bandwidth configured by the current system, that is, the calculated bandwidth and the bandwidth value of the current system If there is a difference of 10% in the amount of change, the average value is used as the initial bandwidth B ₀ to reset the bandwidth;

e. Cycle through step b at intervals of 300 seconds for real-time detection of bandwidth.

Claims (7)

1. the method that detects of the network bandwidth, its characteristic comprises:

A. use default bandwidth to download predetermined data volume, obtain initial bandwidth according to the size of said data volume and the merchant of corresponding download duration from the website of testing the speed;

B. load balancing module and Flow Control module are applied to described initial bandwidth in the control, realize load balancing and Flow Control function;

C. write down the data traffic at least 1 time current 1 second; The current bandwidth average and the current maximum bandwidth that are write down based on described data traffic at every turn; The current bandwidth average of each time asked on average obtain existing bandwidth, the current maximum bandwidth of each time is asked the average current maximum bandwidth average that arrives;

D. described existing bandwidth and current maximum bandwidth average value of averaging are calculated, if the current bandwidth of described mean value and current system configuration poor＞setting threshold then carries out the bandwidth replacement with described mean value as initial bandwidth;

E. the time of setting at interval carries out from the b step cycle and detects in real time to carry out bandwidth.

2. the method that the network bandwidth as claimed in claim 1 detects is characterized by: the data traffic among the step c in the recording occurring continuously 5 times current 1 second, and calculate the current bandwidth average and the current maximum bandwidth of each record according to this data volume of 5 times.

3. the method that the network bandwidth as claimed in claim 1 detects is characterized by: get into steps d after step c repeats 5 times again.

4. the method that the network bandwidth as claimed in claim 1 detects is characterized by: through the existing bandwidth described in the nearest 10 times current bandwidth mean value computation step c; Through the current maximum bandwidth average described in the nearest 10 times current maximum bandwidth calculation procedure c.

5. the method that the network bandwidth as claimed in claim 1 detects is characterized by: according to sample variance the maximum bandwidth average described in the steps d is compensated.

6. the method that detects like the described network bandwidth of one of claim 1 to 5 is characterized by: the setting threshold described in the steps d be current system configuration current bandwidth 10%.

7. the method that detects like the described network bandwidth of one of claim 1 to 5, it is characterized by: the setting-up time described in the step e is 300 seconds.

Images