CN102065022B - Method and device for flow balance of aggregation port, aggregation port and network equipment - Google Patents

Abstract

The invention discloses a method, device, aggregation port and network equipment for flow balance at an aggregation port. The method includes: selecting a flow balancing strategy with the smallest flow difference among member ports within a unit time as the flow balancing strategy used by the aggregation port; Within a unit time, judge whether the aggregation port uses the selected flow balancing strategy, whether the flow difference of each member port becomes larger and the range of change exceeds the set range; if so, re-determine the flow rate of each member port within a unit time The traffic balancing policy with the smallest difference is used as the traffic balancing policy currently used by the aggregation port; if not, the current traffic balancing policy of the aggregation port remains unchanged. The present invention can automatically realize the configuration of the optimal flow balance strategy, and adjust the flow balance strategy in real time when the flow difference between member ports changes, so as to obtain the optimal flow balance strategy after the flow change, so that the flow balance strategy can be The change of flow rate at the aggregation port really plays the role of flow balance.

Description

A method, device, aggregation port and network equipment for flow balancing of aggregation ports

technical field

The invention relates to the field of network communication, in particular to a method, device, aggregation port and network equipment for traffic balance of an aggregation port.

Background technique

Port aggregation refers to the aggregation of multiple ports together to form a virtual aggregation port, so as to realize the sharing of inbound/outbound load among the member ports in the aggregation port, and also provide higher connection reliability. The formed aggregate port is usually called aggregate port (Aggregate Port), and the members forming the aggregate port are called aggregate member ports.

Compared with ordinary ports, the biggest advantage of port aggregation is that it can provide greater bandwidth. As shown in Figure 1, there are 8 100M ports on device A and 8 100M ports on device B. Each connection is logically Only one hundred megabytes of bandwidth can be provided. If the 8 ports of device A form an aggregation port, and the 8 ports of device B also form an aggregation port, it is equivalent to connecting two devices through an aggregation port, which can simulate a gigabit port logically. bandwidth, that is, there are 8 member links to share the traffic of the aggregation port.

The traffic balance configuration on the aggregation port means that the aggregation port can evenly distribute traffic to its member links according to the characteristic field of the packet, so as to avoid traffic concentration on individual member ports.

At present, several commonly used strategies for traffic balancing of aggregation ports include: based on source MAC address, based on destination MAC address, based on source MAC address+destination MAC address, based on source IP address, based on destination IP address, and based on source IP address+destination IP Address and other aggregation port traffic balance policies.

Take the source MAC address-based traffic balancing strategy as an example. Its specific method is to distribute packets to each member port of the aggregation port for forwarding according to the source MAC address of the packet. Packets with different source MAC addresses are forwarded through different member interfaces, and packets with the same source MAC address are forwarded through the same member interface. Through the traffic balance strategy, the traffic on the AP port can be evenly distributed to each member port according to a certain rule, so as to avoid all traffic being forwarded from a certain member port and improve the utilization efficiency of the aggregation port.

Currently, the commonly used traffic balancing policy on the aggregation port can only be pre-configured by the administrator. Once the network traffic changes, the traffic balancing policy on the aggregation port may no longer be optimal, so it cannot achieve a better effect of balancing traffic.

For example, in the application scenario shown in Figure 2, device A and device B are connected through an aggregation port, and device A is also connected to device X and device Y. The aggregation port between device A and device B contains two member ports. Assume that the traffic balancing policy configured by the administrator on the aggregation port between device A and device B is based on the source MAC address. When devices X and Y need to communicate with the external network, assume that the MAC address of device X is 00-00- 00-00-00-01, the MAC address of device Y is 00-00-00-00-00-02, according to the configured traffic balance policy, the traffic of device X is transmitted through link 1, and the traffic of device Y is transmitted through link 1 2 transfers. Once device Y fails and only device X communicates with the external network, all traffic is still transmitted through link 1, which causes a waste of link 2.

Contents of the invention

The embodiment of the present invention provides a flow balancing method and device of an aggregation port, an aggregation port and network equipment, which are used to solve the problem that the flow balancing strategy of the aggregation port in the prior art cannot be adaptively adjusted according to the change of network traffic, resulting in failure to achieve balanced traffic effect problem.

A method for balancing flow at an aggregation port provided in an embodiment of the present invention includes:

Select the traffic balancing policy with the smallest flow difference among member ports per unit time as the traffic balancing policy used by the aggregation port;

Periodically, within a unit time, judge whether the aggregate port uses the selected traffic balance strategy, whether the flow difference of each member port becomes larger and the range of change exceeds the set range;

If so, re-determine the flow balancing strategy with the smallest flow difference among member ports per unit time as the flow balancing strategy currently used by the aggregation port;

If not, maintaining the current traffic balance policy of the aggregation port unchanged;

The flow balance policy with the smallest flow difference among member ports within the unit time is determined by the following method:

For each traffic balancing strategy supported by the aggregation port, calculate the deviation rate of the rate per unit time when the aggregation port uses the traffic balancing strategy per unit time;

The deviation rates obtained by each flow balancing strategy are compared, and the flow balancing strategy with the smallest deviation rate is selected as the flow balancing strategy with the smallest flow difference among member ports within the unit time.

An embodiment of the present invention provides a flow balancing device at a polymerization port, including:

The strategy selection and adjustment module is used to select the flow balance strategy with the smallest flow difference among member ports per unit time as the flow balance strategy used by the aggregation port; The flow balancing strategy with the smallest flow difference is used as the flow balancing strategy currently used by the aggregation port; and when the judgment result is no, the current flow balancing strategy of the aggregation port remains unchanged; the flow difference of each member port in the unit time is the smallest The flow balance strategy is determined by the following method: for each flow balance strategy supported by the aggregation port, calculate the deviation rate of the rate per unit time when the flow balance strategy is used by the aggregation port in a unit time ; Comparing the deviation rates obtained by each flow balancing strategy, selecting the flow balancing strategy with the smallest deviation rate as the flow balancing strategy with the smallest flow difference among member ports within the unit time;

The judging module is configured to, after selecting the traffic balancing strategy with the smallest traffic difference, periodically judge whether the aggregation port uses the selected traffic balancing strategy, whether the traffic difference of each member port becomes larger and The range of change exceeds the set range.

An aggregation port provided by an embodiment of the present invention includes the flow balancing device for an aggregation port provided by an embodiment of the present invention.

A network device provided by an embodiment of the present invention, the network device includes the above aggregation port provided by the embodiment of the present invention.

The beneficial effects of the embodiments of the present invention include:

The method, device, aggregation port, and network equipment provided by the embodiments of the present invention firstly select the flow balancing strategy with the smallest flow difference among member ports per unit time as the traffic balancing strategy used by the aggregation port; then periodically In unit time, judge whether the flow difference of each member port becomes larger and the range of change exceeds the set range when the aggregation port uses the selected flow balancing strategy; if so, re-determine the minimum flow difference of each member port in unit time The traffic balancing policy of the aggregation interface is currently used as the traffic balancing policy, otherwise the current traffic balancing policy remains unchanged. The embodiment of the present invention does not need to pre-configure the traffic balance policy on the aggregation port, and automatically realizes the configuration of the optimal traffic balance policy, and can adjust the traffic balance policy in real time when the traffic difference between the member ports changes, so as to obtain the traffic The changed optimal flow balance strategy enables the flow balance strategy to truly play a role in flow balance as the aggregation port flow changes.

Description of drawings

Fig. 1 is the schematic diagram of aggregation port in the prior art;

FIG. 2 is a schematic diagram of a connection through an aggregation port application scenario in the prior art;

FIG. 3 is a flow chart of a method for balancing flow at an aggregation port provided by an embodiment of the present invention;

FIG. 4 is a schematic connection diagram of an example provided by an embodiment of the present invention;

Fig. 5 is a schematic structural diagram of a device for balancing flow at a polymerization port provided by an embodiment of the present invention.

Detailed ways

The specific implementation manners of a method, device, aggregation port and network equipment provided by the embodiments of the present invention for traffic balancing at an aggregation port will be described in detail below with reference to the accompanying drawings.

The method for balancing flow at an aggregation port provided by the embodiment of the present invention, as shown in FIG. 3 , includes:

S301. Select a traffic balancing policy with the smallest flow difference among member ports per unit time as the traffic balancing policy used by the aggregation port;

S302. Periodically, within a unit time, judge whether the aggregation port uses the selected traffic balancing strategy, whether the flow difference of each member port becomes larger and the range of change exceeds the set range; if so, perform the following step S303; If not, perform the following step S304;

S303. Re-determine the flow balance strategy with the smallest flow difference among member ports within a unit time, and use the re-determined flow balance strategy as the flow balance strategy currently used by the aggregation port, and then turn to step S302;

S304. Keep the current traffic balance policy of the aggregation port unchanged, and then turn to step S302.

The above steps will be described in detail below.

In the above steps S301-S304, at the beginning of the aggregate port’s work, the administrator does not use the method of pre-configuring the traffic balance policy, but according to the various traffic balance policies supported by the aggregate port, each member port of the aggregate port According to the size of the traffic difference, a traffic balancing strategy with the smallest difference is selected as the traffic balancing strategy used by the current aggregation port, and then it is continuously adjusted according to the change of the traffic difference of each member port of the aggregation port obtained in each cycle .

Specifically, the step of selecting the flow balancing strategy with the smallest flow difference among member ports per unit time in the above step S301, and re-determining the flow balance strategy with the smallest flow difference among member ports per unit time in step S303, through the following methods accomplish:

For each traffic balancing strategy supported by the aggregation port, calculate the deviation rate of the rate per unit time when the aggregation port uses the traffic balancing strategy per unit time;

Compare the deviation rates obtained by each flow balancing strategy, and select the flow balancing strategy with the smallest deviation rate as the flow balancing strategy with the smallest flow difference among member ports per unit time.

In the above step S302, when the aggregation port uses the selected traffic balancing strategy, the flow difference of each member port per unit time is also realized by calculating the deviation rate of the aggregation port rate in this period.

Specifically, in the above steps S301-S303, the deviation rate of the aggregation port rate reflects the size of the flow difference between the member ports of the aggregation port. The smaller the flow imbalance between each other, the smaller the flow difference between the member ports of the aggregation port is, and the mutual flow is more balanced. Its calculation method is as follows:

1. Calculate the standard deviation of the rate of each member port of the aggregation port and the sum of the rate of each member port within a unit of time;

2. Divide the standard deviation by the sum of the rates of each member port to obtain the deviation rate of the rate per unit time of the aggregation port.

For example, assuming that there are N member ports in the aggregation port, and the rate of each member port is X ₁ , X ₂ ... X _n in unit time, calculate the rate of each member port in the aggregation port in unit time The formula for the standard deviation of the rate is as follows:

其中：

in:

w is the average value of X ₁ , X ₂ ... X _n ;

X ₁ , X _{2 .} . . X _n are the rates of N member ports in unit time.

in:

为N个成员口速率的总和。

It is the sum of the rates of N member interfaces.

The reason why the standard deviation of the rate of each member port is used to measure the flow difference of each member port is because the standard deviation is used to represent the degree of data dispersion. In the above formula for calculating the rate deviation, the standard deviation of the rate of each member port is divided by The purpose of aggregation interface rate (the sum of the rate of each member interface) is to eliminate the impact of the absolute value of the interface rate, so that aggregation interfaces with different bandwidths can use a unified threshold to measure whether the deviation rate change exceeds the set range. The aggregation port can distribute different traffic to each member port more evenly, so it is necessary to select the traffic balancing strategy with the smallest deviation rate of the rate of each member port as the traffic balancing strategy currently used by the aggregation port.

Preferably, the method for balancing traffic at the aggregation port provided by the embodiment of the present invention further includes the step of saving the deviation rate of the rate per unit time of the selected traffic balancing strategy after the above step S301.

Therefore, in the above-mentioned step S302, if the difference in the flow rate of each member port in the unit time becomes larger and the change range exceeds the set range, it can be judged in the following way: first calculate the deviation rate of the aggregation port rate in the unit time , to judge whether the deviation rate of the polymerization port rate in this unit time is greater than the previously saved deviation rate; if the judgment result is yes, further judge the difference between the deviation rate of the polymerization port rate in this unit time and the saved deviation rate Whether the value is greater than the set first threshold; if so, it can be judged that the difference in the flow rate of each member port within the unit time becomes larger and the range of change exceeds the set range.

Preferably, after the above step S303 re-determines the flow balancing strategy with the smallest flow difference among member ports per unit time and takes it as the flow balancing strategy currently used by the aggregation port, the aggregation port provided by the embodiment of the present invention For the flow balancing method, the following steps can also be performed:

Compare the deviation rate per unit time of the traffic balance strategy re-determined in step S303 with the saved deviation rate; The deviation rate of the rate per unit time of the determined traffic balance strategy updates the saved deviation rate.

If the deviation rate of the re-determined traffic balance policy per unit time is not less than the saved deviation rate, the operation of updating the saved deviation rate is not performed.

The advantage of this is that if the deviation rate of the re-determined traffic balancing policy per unit time is smaller than the saved deviation rate, it means that the current network traffic is changing in a direction that is more in line with the current traffic balancing policy, or using the traffic balancing policy Under the policy, the traffic of each member port of the port tends to be more uniform. If the saved deviation rate is updated at this time, the next adjustment is based on the current updated deviation rate. Since the saved deviation rate is smaller than before, the trigger conditions for the next adjustment of the traffic balance strategy are more stringent (it is conceivable that if the saved deviation rate has a tendency to decrease continuously, the extreme ideal state is The saved deviation rate is zero, that is, as long as the traffic imbalance of each port occurs, it needs to be adjusted accordingly), in the case of the first threshold unchanged, the deviation rate of triggering the traffic balance strategy is reduced, which improves the sensitivity of the adjustment. This makes the adjustment of the traffic balancing policy more flexible.

Preferably, in order to avoid fluctuations caused by abnormal traffic jitter in a short period of time in the network, which continuously triggers related calculations and adjustments of the traffic balancing strategy, in the above method for balancing traffic at the aggregation port provided by the embodiment of the present invention, in S302, each When it is judged that the difference in the flow rate of each member port in the unit time becomes larger and the range of change exceeds the set range, count the number of times that the flow rate difference of each member port in the unit time becomes larger and the range of change exceeds the set range; and Judging whether the number of counts counted in the current period reaches the set second threshold, and when the second threshold is reached, turn to the step of re-determining the flow balancing policy with the smallest flow difference between member ports within a unit time. By setting a specific number of times when the traffic difference changes beyond the set range, when this situation reaches the set threshold, the step of re-determining the traffic balancing strategy with the smallest traffic difference can be initiated, which can avoid abnormal traffic jitter in a short period of time caused vibration problems.

In order to more clearly illustrate the above method of flow balance at the polymerization port provided by the embodiment of the present invention, a simple example is used below to illustrate:

As shown in Figure 4, device A and device B are connected through an aggregation port, which consists of two member ports. 3 devices X, Y, Z and A are connected, 2 devices M, N and device B are connected. Assume that device A supports only source MAC-based and destination MAC-based traffic balancing policies. Assume that the source MAC addresses of devices X, Y, and Z are 00-00-00-00-00-01, 00-00-00-00-00-02, and 00-00-00-00-00-03, respectively. Assume that the source MAC addresses of devices M and N are 01-23-45-00-00-01 and 01-23-45-00-00-02 respectively. The preset unit time on device A is 1 minute, and the first threshold is 0.2.

The selection and adjustment process of aggregate port traffic balance policy is described as follows:

Assume that at the beginning, devices X, Y, and Z send 100 packets to devices M and N respectively every minute, then the traffic of the aggregation port per minute is 600 packets.

1. In order to select the traffic balancing strategy with the smallest flow difference among member ports, the aggregation port on device A first selects the first balancing strategy as the traffic balancing strategy based on the source MAC, and the aggregation port uses this strategy to transfer the traffic sent by X device and Z device. The packet is sent from link 1, and the packet sent by the Y device is sent from link 2. When the first minute expires, count the number of packets sent by one member port port1 as 400, and the number of packets sent by the other member port port2 as 200. It is calculated that the standard deviation of the rate under the traffic balance strategy is 100, and the deviation rate S is 0.16.

2. In the second minute, the aggregation port on device A selects the second traffic balancing policy as the traffic balancing policy based on the destination MAC. N's message is sent from link 2. When the second minute expires, the number of packets sent on the member port port1 is counted as 300, and the number of packets sent on the other member port port2 is 300. It is calculated that the rate standard deviation is 0 and the deviation rate S is 0 under the traffic balance strategy.

3. By comparing the deviation rates obtained by the two units of time, the aggregation port on device A selects a traffic balancing strategy based on the destination MAC for application, and saves the current deviation rate S=0.

Afterwards, every minute, the deviation rate S` is calculated on device A. If the calculated deviation rate S` becomes smaller or larger but the change range is less than 0.2, the traffic balance policy applied on the aggregation port will not be modified. Otherwise, it will need to Re-elect the traffic balancing policy.

4. Assuming that device N fails and shuts down during operation, the network traffic becomes X, Y, and Z sends 100 messages to device M every minute.

At this time, when device A calculates the rate standard deviation, since the destination MAC address of all packets is M, it only transmits on link 1. According to statistics, the number of packets sent on the member port port1 is 300, and the number of packets sent on the other member port port2 is 0. It is calculated that under the traffic balance strategy based on the destination MAC, the standard deviation of the rate of each member port is 150, and the deviation rate is S`=0.5. Compared with S`, which becomes larger than S and the range of change exceeds 0.2, it is necessary to re-elect the traffic balancing strategy.

The traffic balance policy selected for the aggregation port on device A is based on the source MAC address. When one minute expires, the number of packets sent on port1 is counted as 200, and the number of packets sent on port2 is 100. It is calculated that the rate variance under the traffic balance strategy is 50, and the deviation rate S is 0.16.

Compared with the two supported traffic balancing policies, the deviation rate of the traffic balancing policy based on the source MAC is small, and the traffic balancing policy of the aggregation port is changed to the traffic balancing policy based on the source MAC.

Based on the same inventive concept, the embodiment of the present invention also provides a device for balancing traffic at an aggregated port, an aggregated port, and network equipment. Since the problem-solving principles of the device, the aggregated port, and the network device are the same as those of the aforementioned method for balancing traffic at the aggregated port Similar, therefore, the implementation of these devices, aggregation ports, and network equipment can refer to the implementation of the aforementioned methods, and repeated descriptions will not be repeated.

The device for balancing flow at the aggregation port provided by the embodiment of the present invention, as shown in FIG. 5 , includes:

The strategy selection and adjustment module 501 is used to select the flow balance strategy with the smallest flow difference among member ports per unit time as the flow balance strategy used by the aggregation port; The traffic balancing strategy with the smallest port flow difference is used as the traffic balancing strategy currently used by the aggregation port; and when the judgment result is no, the current flow balancing strategy of the aggregation port remains unchanged;

The judging module 502 is configured to, after selecting the traffic balancing strategy with the smallest traffic difference, periodically and within a unit time, judging whether the traffic difference of each member port becomes larger when the aggregation port uses the selected traffic balancing strategy And the range of change exceeds the set range.

Further, the above-mentioned policy selection and adjustment module 501 is further configured to calculate, for each traffic balancing strategy supported by the aggregation port, the traffic balance strategy used by the aggregation port in a unit time, in the unit time The deviation rate of the rate; compare the deviation rates obtained by each flow balancing strategy, and select the flow balancing strategy with the smallest deviation rate as the flow balancing strategy with the smallest flow difference among member ports within the unit time.

Further, the above-mentioned strategy selection and adjustment module 501 is specifically used to calculate the standard deviation of the rate of each member port of the aggregation port and the sum of the rates of each member port within a unit time; divide the standard deviation by the sum , to obtain the rate of deviation of the polymerization port per unit time rate.

Further, the device for traffic balancing at the aggregation port provided by the embodiment of the present invention, as shown in FIG. 5 , further includes: a storage module 503, which is used to store the rate of the traffic balancing strategy selected by the policy selection and adjustment module in unit time. deviation rate.

Correspondingly, the above-mentioned judging module 502 is further used to calculate the deviation rate of the aggregation port rate in this unit time; judge whether the deviation rate of the aggregation port rate in the unit time is greater than the stored deviation rate; if the judgment result is yes In the case of , it is further judged whether the difference between the deviation rate of the aggregation port rate within the unit time and the stored deviation rate is greater than the set first threshold; becomes large and the range of change exceeds the setting range.

Preferably, the storage module in the device for balancing aggregated port traffic provided by the embodiment of the present invention is further configured to re-determine the traffic balancing strategy with the smallest traffic difference among member ports per unit time, and then rebalance the re-determined traffic. The policy deviation rate per unit time is compared with the saved deviation rate; when the saved deviation rate is relatively large, the saved deviation rate is updated using the re-determined traffic balance policy rate deviation rate per unit time.

The above-mentioned aggregation port flow balance device provided by the embodiment of the present invention can be implemented in a hardware or a combination of hardware and software. For example, it can be designed as a module independent of each member port of the aggregation port. The balancing device is connected to each member port of the aggregation port, monitors and calculates the flow difference of each member port under the various flow balancing strategies used by the aggregation port, and adjusts the flow balancing strategy of the aggregation port in due course.

Correspondingly, an embodiment of the present invention also provides an aggregation port, which may include the above-mentioned device for balancing traffic at the aggregation port provided by the embodiment of the present invention in addition to each member port.

Furthermore, an embodiment of the present invention also provides a network device, which includes the above aggregation port provided by the embodiment of the present invention.

The method, device, aggregation port, and network equipment provided by the embodiments of the present invention firstly select the flow balancing strategy with the smallest flow difference among member ports per unit time as the traffic balancing strategy used by the aggregation port; then periodically In unit time, judge whether the flow difference of each member port becomes larger and the range of change exceeds the set range when the aggregation port uses the selected flow balancing strategy; if so, re-determine the minimum flow difference of each member port in unit time The traffic balancing policy of the aggregation interface is currently used as the traffic balancing policy, otherwise the current traffic balancing policy remains unchanged. The embodiment of the present invention does not need to pre-configure the traffic balance policy on the aggregation port, and automatically realizes the configuration of the optimal traffic balance policy, and can adjust the traffic balance policy in real time when the traffic difference between the member ports changes, so as to obtain the traffic The changed optimal flow balance strategy enables the flow balance strategy to truly play a role in flow balance as the aggregation port flow changes.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and equivalent technologies thereof, the present invention also intends to include these modifications and variations.

Claims (13)

1. the method for an aggregation port flow equilibrium is characterized in that, comprising:

The flow equilibrium strategy of each member port flow difference minimum is as the employed flow equilibrium strategy of aggregation port in selection unit's time;

In the unit interval, judge this aggregation port under the situation of using selected flow equilibrium strategy periodically, whether each member port flow difference becomes big and amplitude of variation exceeds the scope of setting;

If the flow equilibrium strategy of confirming each member port flow difference minimum in the unit interval again is as the presently used flow equilibrium strategy of said aggregation port;

If not, it is constant to keep said aggregation port present flow rate equilibrium strategy;

The minimum flow equilibrium strategy of each member port flow difference in the said unit interval, confirm through following method:

To each flow equilibrium strategy that said aggregation port is supported, under the situation of unit of account said this flow equilibrium strategy of aggregation port use in the time, the deviation ratio of speed in the unit interval;

Every kind of resulting said deviation ratio of flow equilibrium strategy is compared, select the minimum flow equilibrium strategy of said deviation ratio as the minimum flow equilibrium strategy of each member port flow difference in the said unit interval.

2. the method for claim 1 is characterized in that, calculates the deviation ratio of speed in the aggregation port unit interval, comprising:

In the unit interval, calculate the standard deviation of the speed of this each member port of aggregation port, and the summation of each member port speed;

Said standard deviation divided by said summation, is obtained the deviation ratio of speed in this aggregation port unit interval.

3. according to claim 1 or claim 2 method is characterized in that, select the minimum flow equilibrium strategy of deviation ratio as the said unit interval in after the minimum flow equilibrium strategy of each member port flow difference, also comprise:

Preserve the deviation ratio of selected flow equilibrium strategy speed in the unit interval.

4. method as claimed in claim 3 is characterized in that, said should the unit interval in the big and amplitude of variation of each the member port difference in flow mutation scope that exceeds setting judge through following manner:

Calculate the deviation ratio of this aggregation port speed in this unit interval;

Whether the deviation ratio of judging aggregation port speed in this unit interval is greater than the said deviation ratio of preserving;

In judged result is under the situation that is, whether the difference between the deviation ratio of further judging aggregation port speed in this unit interval and the said deviation ratio of preservation is greater than the first threshold of setting; If, then judge should be in the unit interval the big and amplitude of variation of each member port difference in flow mutation exceeded setting range.

5. method as claimed in claim 3 is characterized in that, confirms again also to comprise after the flow equilibrium strategy of each member port flow difference minimum in the unit interval:

The deviation ratio of said flow equilibrium strategy speed in the unit interval of confirming again and the deviation ratio of preservation are compared;

When the deviation ratio of preserving is big, use the deviation ratio of said flow equilibrium strategy speed in the unit interval of confirming again to upgrade the deviation ratio of preserving.

6. according to claim 1 or claim 2 method is characterized in that, when judging in this unit interval that the big and amplitude of variation of each member port difference in flow mutation exceeds the scope of setting, also comprises:

The number of times of situation that the big and amplitude of variation of each member port difference in flow mutation in unit interval exceeds the scope of setting appears in statistics;

Judge whether the number of times of current period statistics reaches second threshold value of setting, and when reaching said second threshold value, turn to execution to confirm the step of the flow equilibrium strategy that member port flow difference in the unit interval is minimum again.

7. the device of an aggregation port flow equilibrium is characterized in that, comprising:

Policy selection and adjusting module, the flow equilibrium strategy that is used for each member port flow difference minimum in selection unit's time is as the employed flow equilibrium strategy of aggregation port; And in the judge module judged result when being, confirm that again the minimum flow equilibrium strategy of each member port flow difference in the unit interval is as the presently used flow equilibrium strategy of said aggregation port; And in judged result for not the time, it is constant to keep aggregation port present flow rate equilibrium strategy; The minimum flow equilibrium strategy of each member port flow difference in the said unit interval; Confirm through following method: be directed against each flow equilibrium strategy that said aggregation port is supported; Under the situation of unit of account said this flow equilibrium strategy of aggregation port use in the time, the deviation ratio of speed in the unit interval; Every kind of resulting said deviation ratio of flow equilibrium strategy is compared, select the minimum flow equilibrium strategy of said deviation ratio as the minimum flow equilibrium strategy of each member port flow difference in the said unit interval;

Judge module; Be used for after selecting the minimum flow equilibrium strategy of said flow difference; In the unit interval, judge this aggregation port under the situation of using selected flow equilibrium strategy periodically, whether each member port flow difference becomes big and amplitude of variation exceeds the scope of setting.

8. device as claimed in claim 7 is characterized in that, said policy selection and adjusting module specifically are used in the unit interval, calculating the standard deviation of the speed of this each member port of aggregation port, and the summation of each member port speed; Said standard deviation divided by said summation, is obtained the deviation ratio of speed in this aggregation port unit interval.

9. like claim 7 or 8 described devices, it is characterized in that, also comprise: memory module is used to store the deviation ratio of said policy selection and the selected flow equilibrium strategy of adjusting module speed in the unit interval.

10. device as claimed in claim 8 is characterized in that, said judge module is further used for calculating the deviation ratio of this aggregation port speed in this our unit time; Whether the deviation ratio of judging aggregation port speed in this unit interval is greater than the said deviation ratio of preserving; In judged result is under the situation that is, whether the difference between the deviation ratio of further judging aggregation port speed in this unit interval and the said deviation ratio of preservation is greater than the first threshold of setting; If, then judge should be in the unit interval the big and amplitude of variation of each member port difference in flow mutation exceeded setting range.

11. device as claimed in claim 9; It is characterized in that; Said memory module; Also be used for after the flow equilibrium strategy of confirming each member port flow difference minimum in the unit interval again, the deviation ratio of said flow equilibrium strategy speed in the unit interval of confirming again and the deviation ratio of preservation being compared; When the deviation ratio of preserving is big, use the deviation ratio of said flow equilibrium strategy speed in the unit interval of confirming again to upgrade the deviation ratio of preserving.

12. an aggregation port is characterized in that said aggregation port comprises the device like each described aggregation port flow equilibrium of claim 7-11.

13. a network equipment is characterized in that the said network equipment comprises aggregation port as claimed in claim 12.

Images