English to French Tech/Engineering Patents METHOD AND SYSTEM FOR NETWORK CONGESTION MANAGEMENT

In step S401: when the bandwidth restoring flow is over or no congestion notification message is received within a set period of time (the duration T2 for weighing feasibility of probing, which may be confïgured and dépends on the frequency of bandwidth probing expectêd by the user) during normal opération of the network and the data transmission rate needs to be increased, triggering to enter a bandwidth probing stage, i.e., entering the part of increasing the bandwidth utilization ratio in the method for network congestion management.
In step S402: tentatively increasing the current data transmission rate to probe the current available bandwidth, and resetting the timer/counter T2 for bandwidth probing.
In step S403: judging whether a congestion notification message is received during the opération period after the timer/counter T2 is reset and before it overflows, if yes (the network has congestion or the probing fails), going to the flow of congestion processing and bandwidth restoring after réception of the congestion notification message (i.e., step S302) in the method for congestion management shown in FIG. 3; if not (the network is normal or the probing succeeds), execùting step S404 sequentially.
Step S404: judging whether the timer/counter for weighing probing overflows, if yes, going to step S402 to probe available bandwidth again; otherwise, going to step S403 to continue to monitor the situation of network opération, i.e., monitoring whether network is congested.

Of course, the step S401 may be executed when it is judged that the timer/counter overflows in the step S404 of judging whether the timer/counter for weighing probing overflows so that the upstream network entity only increases the data transmission rate to the desired rate and then stops increasing, which will make the network more stable.

In addition, when the network opérâtes normally, there may be much potential bandwidth in the network due to decrease of service amount in a certain period of time or in a certain région, therefore, it is very feasible to probe the network bandwidth.

After the bandwidth of the network entity is completely restored, â bandwidth probing stage is started according to practical requirements, which may be implemented by increasing the current rate by a spécifie constant value. This value may be configured by the user freely, and in this example, this value is supposed to be lGbps. As shown in FIG. 7, the rate is increased to be approximate to 1 lGbps after the first probing, and if this probing does hot cause congestion, then the second probing is implemented to adjust the current rate to 12Gbps, and so forth. When the rate is increased to a certain extent to cause network congestion, the congestion management flow will be started and the rate will be decreased appropriately according to the extent of the congestion; when the network opérâtes normally, and the timer/counter for weighing feasibility of bandwidth probing expires, the network entity is triggered to enter a bandwidth probing stage to increase the current rate by a spécifie constant value; circulation may be made repeatedly to progressively increase the data transmission rate to improve the bandwidth utilization ratio; the détails are as described above and will not be repeated hère.

The data transmission rate adjusting unit 803 may comprise: a rate limiting sub-unit 8031, a data transmission rate restoring sub-unit 8032 and a bandwidth probing sub-unit 8033.
The bandwidth probing sub-unit 8033 is used to increase the data transmission rate of the upstream network entity by a certain step length when the upstream network entity transmits data at a rate greater than equal to the normal rate value for a certain period of time and the data transmission rate needs to be increased; the détails are as described above.