WO2017049925A1

WO2017049925A1 - Network congestion control method and apparatus, and client

Info

Publication number: WO2017049925A1
Application number: PCT/CN2016/082528
Authority: WO
Inventors: 赵洋
Original assignee: 中兴通讯股份有限公司
Priority date: 2015-09-22
Filing date: 2016-05-18
Publication date: 2017-03-30
Also published as: CN106549876A

Abstract

A network congestion control method and apparatus, and a client. The network congestion control method comprises: acquiring a delay for representing a network state; comparing the delay with a pre-set delay threshold; and controlling a sending strategy of a client according to a comparison result.

Description

Network congestion control method, device and client

Technical field

This application relates to, but is not limited to, the field of communication technology.

Background technique

Network congestion refers to a situation in which the network transmission performance is degraded due to the limited resources of the store-and-forward node when the number of packets transmitted in the packet-switched network is too large. When the network is congested, data loss, delay increases, and throughput drop are generally caused. In severe cases, "congestion collapse" may occur. There are several main factors that cause network congestion: storage space limitations, bandwidth capacity limitations, and processor performance limitations. In the related art, the client does not know the processing capability of the current transmission network environment or the server, and always sends the request message according to its maximum processing capability. When the transmission network condition is poor or the processing capability of the server is weak, a large number of request messages are generated. Congestion in the network or squeezed on the server side further deteriorates the network environment or crushes the server. Therefore, a method capable of link congestion control is urgently needed.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

The present invention provides a network congestion control method, device, and client to solve the related art. Since the client does not understand the processing network environment or the processing capability of the server, the information is always exchanged with the server according to a single data transmission policy. The technical problem that causes the network environment to deteriorate.

A network congestion control method includes:

Obtain a delay for reflecting the state of the network;

Comparing the delay with a preset delay threshold;

Control the client's sending policy based on the comparison result.

Optionally, in an embodiment of the present invention, the acquired delay is a transmission delay that represents a transmission network environment, a processing delay that characterizes a processing capability of the server, or an overall delay that characterizes the network environment.

Optionally, in an embodiment of the present invention, when the obtained delay is the transmission delay, the manner of obtaining the transmission delay includes one of the following three types:

Obtaining, by the client, a time for sending the first message to the server as a first time point, acquiring a time when the server receives the first message as a second time point, and using the second time point The difference between the first time points is used as the transmission delay;

Obtaining, by the server, a time for sending the second message to the client as a third time point, acquiring a time when the client receives the second message as a fourth time point, and using the fourth time point The difference between the third time points is used as the transmission delay;

Obtaining, by the client, a time for sending the first message to the server as a first time point, acquiring a time when the server receives the first message as a second time point, and using the second time point The difference between the first time point is used as the first transmission delay; the time when the server sends the second message to the client is used as the third time point, and the client receives the second message. The time is taken as the fourth time point, and the difference between the fourth time point and the third time point is used as the second transmission delay, and the sum of the first transmission delay and the second transmission delay is calculated as The transmission delay.

Optionally, in an embodiment of the present invention, when the obtained delay is the processing delay, acquiring the processing delay includes:

Obtaining, as the second time point, the time when the server receives the first message sent by the client, and acquiring the time when the server sends the second message to the client, as the third time point, where the The difference between the three time points and the second time point is taken as the processing delay.

Optionally, in an embodiment of the present invention, when the obtained delay is the overall time delay, the manner of obtaining the total delay includes one of the following three types:

Obtaining, as the first time, the time when the client sends the first message to the server, and acquiring the time when the server sends the second message to the client as the third time point, where the third time is a difference between the point and the first time point as the overall time delay;

Obtaining, as the second time point, the time when the server receives the first message sent by the client, and acquiring the time when the client receives the second message sent by the server, as the fourth time point, Determining a difference between the fourth time point and the second time point as the overall time delay;

Obtaining, as the first time, the time when the client sends the first message to the server And taking the time when the client receives the second message sent by the server as the fourth time point, and using the difference between the fourth time point and the first time point as the overall time delay.

Optionally, in an embodiment of the present invention, the preset delay threshold includes a first threshold and a second threshold, and the first threshold is smaller than the second threshold.

Optionally, in an embodiment of the present invention, the controlling a sending policy of the client according to the comparison result includes:

When the comparison result is that the delay is greater than or equal to the second threshold, reducing a frequency at which the client sends a message to the server;

And maintaining, when the comparison result is that the delay is greater than or equal to the first threshold and less than the second threshold, maintaining a frequency at which the client sends a message to the server;

And when the comparison result is that the delay is less than the first threshold, maintaining a frequency at which the client sends a message to the server or increasing a frequency at which the client sends a message to the server.

When the comparison result is that the delay is greater than or equal to the second threshold, determining whether the first N-1 comparison results before the current comparison result are all the delay is greater than or equal to the second threshold; If the result of the determination is yes, the frequency at which the client sends a message to the server is decreased, and when the determination result is no, the frequency at which the client sends a message to the server is maintained;

When the comparison result is that the delay is less than the first threshold, determining whether the first N-1 comparison results before the current comparison result are all the delay is less than the first threshold; when the judgment result is yes The frequency of the message sent by the client to the server is increased, and when the determination result is no, the frequency at which the client sends a message to the server is maintained.

A network congestion control device includes:

The delay acquisition module is configured to: obtain a delay for reflecting a network state;

The delay comparison module is configured to: compare the delay acquired by the delay acquisition module with a preset delay threshold;

The policy control module is configured to: control a sending policy of the client according to the comparison result obtained by the delay comparison module.

Optionally, in an embodiment of the present invention, the delay obtaining module includes at least one of a transmission delay acquisition submodule, a processing delay acquisition submodule, and an overall delay acquisition submodule;

The transmission delay acquisition submodule is configured to: acquire a transmission delay that characterizes a transmission network environment;

The processing delay acquisition submodule is configured to: acquire a processing delay that represents a processing capability of the server;

The overall delay acquisition submodule is configured to: obtain an overall delay characterizing a network environment

Optionally, in an embodiment of the present invention, when the time delay acquisition module includes the transmission delay acquisition submodule, the transmission delay acquisition submodule includes:

The first obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a second acquiring unit, configured to: acquire a time when the server receives the first message as a second time point;

a first calculating unit, configured to: calculate, as the transmission delay, a difference between the second time point acquired by the second acquiring unit and the first time point acquired by the first acquiring unit;

Alternatively, the transmission delay acquisition submodule includes:

The third obtaining unit is configured to: acquire a time when the server sends the second message to the client as a third time point;

a fourth obtaining unit, configured to: acquire a time when the client receives the second message as a fourth time point;

a second calculating unit, configured to: calculate, as the transmission delay, a difference between the fourth time point acquired by the fourth acquiring unit and the third time point acquired by the third acquiring unit;

Alternatively, the transmission delay acquisition submodule includes:

The fifth obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a sixth obtaining unit, configured to: acquire a time when the server receives the first message as a second time point;

The seventh obtaining unit is configured to: acquire a time when the server sends the second message to the client as a third time point;

An eighth obtaining unit, configured to: acquire a time when the client receives the second message as a fourth time point;

a third calculating unit, configured to: calculate a difference between the second time point acquired by the sixth acquiring unit and the first time point acquired by the fifth acquiring unit as a first transmission delay, and calculate the a difference between the fourth time point acquired by the eighth obtaining unit and the third time point acquired by the seventh acquiring unit as a second transmission delay, and calculating the first transmission delay and the second The sum of transmission delays is taken as the transmission delay.

Optionally, in an embodiment of the present invention, when the time delay obtaining module includes the processing delay acquiring submodule, the processing delay obtaining submodule includes:

The ninth obtaining unit is configured to: acquire, as the second time point, the time when the server receives the first message sent by the client;

The tenth acquiring unit is configured to: acquire, as the third time point, a time when the server sends the second message to the client;

The fourth calculating unit is configured to: calculate, as the processing delay, a difference between the third time point acquired by the tenth acquiring unit and the second time point acquired by the ninth acquiring unit.

Optionally, in an embodiment of the present invention, when the time delay acquisition module includes the total delay acquisition submodule, the total delay acquisition submodule includes:

The eleventh acquiring unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a twelfth obtaining unit, configured to: acquire a time when the server sends a second message to the client as a third time point;

a fifth calculating unit, configured to: calculate, as the overall delay, a difference between the third time point acquired by the twelfth acquiring unit and the first time point acquired by the eleventh acquiring unit;

Alternatively, the overall delay acquisition submodule includes:

a thirteenth obtaining unit, configured to: acquire a time when the server receives the first message sent by the client as a second time point;

a fourteenth obtaining unit, configured to: acquire a time when the client receives the second message sent by the server as a fourth time point;

a sixth calculating unit, configured to: calculate, as the overall delay, a difference between the fourth time point acquired by the fourteenth acquiring unit and the second time point acquired by the thirteenth acquiring unit;

Alternatively, the overall delay acquisition submodule includes:

The fifteenth obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a sixteenth obtaining unit, configured to: acquire a time when the client receives the second message sent by the server as a fourth time point;

The seventh calculating unit is configured to: calculate, as the overall delay, a difference between the fourth time point acquired by the sixteenth acquiring unit and the first time point acquired by the fifteenth acquiring unit.

Optionally, in an embodiment of the present invention, the delay comparison module includes a first threshold comparison submodule and a second threshold comparison submodule;

The first threshold comparison submodule is configured to: compare the time delay acquired by the delay acquisition module with a first threshold;

The second threshold comparison submodule is configured to compare the time delay acquired by the delay acquisition module with a second threshold.

Optionally, in an embodiment of the present invention, the policy control module controls a sending policy of the client according to the comparison result obtained by the delay comparison module, including:

When the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the second threshold, the policy control module is configured to: reduce a frequency at which the client sends a message to the server;

When the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the first threshold and less than the second threshold, the policy control module is configured to: maintain the client to the The frequency at which the server sends messages;

When the comparison result obtained by the delay comparison module is that the delay is less than the first threshold, the policy control module is configured to: maintain a frequency or improve the frequency at which the client sends a message to the server. The frequency at which the client sends a message to the server.

Optionally, in an embodiment of the present invention, the policy control module further includes a determining submodule and a control submodule:

The determining sub-module is configured to: when the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the second threshold, determine whether the first N-1 comparison results before the current comparison result are all The control sub-module is configured to: when the judgment result of the determining sub-module is YES, reduce the frequency at which the client sends a message to the server, when And when the determining result of the determining submodule is negative, maintaining a frequency at which the client sends a message to the server;

The control sub-module is further configured to: when the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the first threshold and less than the second threshold, maintaining the client to the service The frequency at which the message is sent by the terminal;

The determining sub-module is further configured to: when the comparison result obtained by the delay comparison module is that the delay is less than the first threshold, determine whether the first N-1 comparison results before the current comparison result are all The control sub-module is further configured to: when the determination result of the determining sub-module is YES, increase the frequency at which the client sends a message to the server, when When the judgment result of the submodule is negative, the frequency at which the client sends a message to the server is maintained.

The present invention also provides a client comprising the network congestion control device according to any of the above.

The network congestion control method, device, and client provided by the embodiment of the present invention obtain the time delay of the network state, and compare the acquired delay with the preset delay threshold to learn the current network condition, and then according to the current network situation. The comparison result controls the sending policy of the client, achieves the real-time understanding of the current network state, and adjusts the data transmission policy of the client according to the current network state, and solves the problem that the processing capability of the network environment or the server is not understood by the client in the related technology. It always interacts with a single data transmission strategy, which leads to the deterioration of the network environment, improves the efficiency of client-server interaction, and optimizes network resource configuration.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

BRIEF abstract

FIG. 1 is a flowchart of a network congestion control method according to Embodiment 1 of the present invention;

2 is a schematic structural diagram of a network congestion control apparatus according to Embodiment 2 of the present invention;

3 is a schematic structural diagram of a delay acquisition module in a network congestion control apparatus provided in the embodiment shown in FIG. 2;

4 is a schematic structural diagram of a transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

5 is a schematic structural diagram of another transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

6 is a schematic structural diagram of another transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

7 is a schematic structural diagram of a processing delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

8 is a schematic structural diagram of an overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

FIG. 9 is a schematic structural diagram of another overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

10 is a schematic structural diagram of another overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3;

FIG. 11 is a schematic structural diagram of a delay comparison module in a network congestion control apparatus according to the embodiment shown in FIG. 2;

12 is a schematic structural diagram of a policy control module in a network congestion control apparatus provided in the embodiment shown in FIG. 2;

FIG. 13 is a schematic structural diagram of another policy control module in the network congestion control apparatus provided in the embodiment shown in FIG. 2.

Embodiments of the invention

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted, The features in the embodiments and the embodiments herein may be arbitrarily combined with each other without conflict.

The steps illustrated in the flowchart of the figures may be executed in a computer system such as a set of computer executable instructions. Also, although logical sequences are shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.

Typically, network congestion occurs primarily due to storage space and bandwidth capacity limitations and processor performance limitations. Therefore, in theory, solving network congestion can be done from three aspects: increasing storage space and bandwidth capacity or improving processor performance. However, in reality, storage space and bandwidth capacity cannot be increased indefinitely, and processor performance is also improved. It is impossible to increase infinitely, so the most practical way to solve network congestion is to perform network congestion control. The network congestion control method and device and the client provided by the embodiment of the present invention obtain a delay capable of transmitting a network environment or a server processing capability, compare the delay with a preset delay threshold, and then control the client according to the comparison result. The data transmission strategy is to achieve real-time understanding of the current network status, and adjust the client's data transmission policy according to the current network status. The present invention will be further described in detail below by way of specific embodiments with reference to the accompanying drawings:

Embodiment 1:

FIG. 1 is a flowchart of a network congestion control method according to Embodiment 1 of the present invention. The network congestion control method provided in this embodiment may include steps 101 to 103:

Step 101: Obtain a delay for reflecting a network status.

In this embodiment, the obtained delay may be a transmission delay, a processing delay, or an overall delay.

The transmission delay can be the time taken by the client to interact with the server in the transmission process, so the transmission delay can characterize the transmission network. On the one hand, when the client sends a message to the server, the sending time of the message is intercepted as the first time point, and the time when the server receives the message is taken as the second time point, and the second time point and the first time point are The time difference between the two is the transmission delay, and the delay between the second time point and the first time point is only a one-way transmission delay. On the other hand, another one-way transmission delay can also be obtained. When the server sends a message to the client, the sending time of the message is intercepted as the third time point, and the time when the client receives the message is taken as the fourth time. At the time point, the time difference between the fourth time point and the third time point can also be used as the transmission delay. again In terms of the two-way transmission delay, the sum of the two-way transmission delay is one of the transmission delays.

The processing delay can be the time taken by the server to process the message sent by the client, that is, the time between receiving the request message from the server and the response message sent by the server to the client, and the processing time reflects the processing capability of the server. . When the client sends a message to the server, the interception server receives the request message sent by the client as the second time point, and intercepts the time when the server sends the response message to the client as the third time point, and compares the first time. The second time point and the third time point are obtained, and the time difference between the third time point and the second time point is obtained as a processing delay.

There are many ways to obtain the overall delay. In general, the overall delay acquisition method can be divided into indirect acquisition and direct acquisition:

Since the overall delay is the sum of the transmission delay and the processing delay, it corresponds to the three forms of obtaining the transmission delay, and the indirect acquisition methods of the overall delay mainly have the following three types:

The first method is: obtaining the time when the client sends the request message to the server as the first time point, and obtaining the time when the server receives the request message as the second time point, and the difference between the second time point and the first time point As the first transmission delay, the time when the server sends the response message to the client is taken as the third time point, and the difference between the third time point and the second time point is taken as the processing delay; the first transmission delay and the processing are calculated. The sum of delays as the overall delay;

The second method is to obtain the time when the server sends a response message to the client as the third time point, and obtain the time when the client receives the response message as the fourth time point, and the difference between the fourth time point and the third time point. As the second transmission delay, the time when the acquiring server receives the request message sent by the client is used as the second time point, and the difference between the third time point and the second time point is used as the processing delay; and the second transmission delay is calculated. And the sum of the processing delays as the overall delay;

The third method is: obtaining the time when the client sends the request message to the server as the first time point, and obtaining the time when the server receives the request message as the second time point, and taking the difference between the second time point and the first time point as The first transmission delay is obtained as the third time point when the server sends the response message to the client, and the time when the client receives the response message is taken as the fourth time point, and the fourth time point and the third time point are The difference is used as the second transmission delay, and the sum of the first transmission delay and the second transmission delay is calculated as the transmission delay, and the difference between the third time point and the second time point is used as the processing delay, and the transmission delay is calculated. And the sum of processing delays as the overall delay.

Optionally, the method for directly obtaining the overall delay may be adopted in this embodiment. This method is simple, and the intermediate link of indirect acquisition is omitted. The direct acquisition method of the overall delay mainly includes the following three types:

The first method is: obtaining the time when the client sends the request message to the server is the first time, and the time for the server to send the response message to the client is the third time point, and the third time point and the first time point are directly The difference is the way of overall delay;

The second mode: the time when the server receives the request message sent by the client is the second time point, and the time when the client receives the response message sent by the server is the fourth time point, and then the fourth time point is directly The difference from the second time point is the overall delay;

The third method: obtaining the time when the client sends the request message to the server is the first time point, and obtaining the response time sent by the client to the server is the fourth time point, and then directly the fourth time point and the The difference in time is used as the overall delay.

When the delay is obtained, the time stamp may be set in the information exchanged between the client and the server, that is, the extension field is added in at least one of the request message or the response message, and the client and the server are added. The party that sends the information records the timestamp of the sending request message or the response message in the extension field. When the other party receives the information, it compares the timestamp contained in the information with its own system time, and calculates that the message is transmitted. The delay in the process. For example, in an embodiment of the present invention, when sending a request message to a server, the client may set a timestamp of the sending time in the request message, and when the server receives the request message, obtain a timestamp from the server. Compare with your own system time to obtain the transmission delay of the request message.

In the actual application, the method of setting the timestamp is not limited to the foregoing manner. For example, the client may record the time when the request message is sent to the server, but the timestamp of the time is not set in the request message. When the server receives the above request message, it records the received time, and sets the timestamp of the time in the response message of the replying client. When the client receives the response message replied by the server, the response message is received from the response message. Obtain the timestamp set by the server and compare it with the time of the previous record to obtain the transmission delay of the above request message.

The above method is set by one of the client or the server, and the other party calculates the delay, and the system time of the client and the server must be synchronized.

Optionally, in this embodiment, the party that sets the timestamp is the party that calculates the delay. For example, when the client sends the request message to the server, when the sending time is set in the request message, The timestamp, when the server receives the request message and sends a response message, the timestamp is copied into the extension field of the response message, and when the client receives the response message sent by the server, the timestamp is obtained. Compared with your own system time, you can calculate the sum of transmission delay and processing delay to reflect the overall network status. This method of obtaining the delay is not required to set the time stamp by one of the client or the server, and the other party does not need to require the system time synchronization between the server and the client.

Optionally, the present invention further provides an embodiment, the delay is obtained by setting a counter, and the client may start to use the counter to start timing when sending the request message to the server, when receiving the response message from the server. Stop timing, this method is simple, there is no need to add extension fields in the interaction information between the client and the server, and no calculation is needed.

Step 102: Compare the acquired delay with a preset delay threshold.

The result of the comparison between the delay and the preset delay threshold may reflect the current network state. Optionally, in this embodiment, two preset thresholds may be set, that is, the preset delay threshold includes the first threshold and the first threshold. The second threshold is that the first threshold is less than the second threshold. When the value of the delay is greater than or equal to the second threshold, the current network state is bad. When the value of the delay is greater than or equal to the first threshold and less than the second threshold, Indicates that the current network status is normal. When the value of the above delay is less than the first threshold, the current network status is good.

The value of the preset delay threshold can be set by the user, and the user can flexibly set according to the storage space and bandwidth capacity of the network used by the user and the performance of the processor. Optionally, in practical applications, the value of the first threshold may be equal to the value of the second threshold.

Step 103: Control a sending policy of the client according to the comparison result.

After obtaining the comparison result according to the above steps, the current client's transmission policy may be controlled according to the comparison result. Optionally, when the comparison result is that the delay is greater than or equal to the second threshold, the frequency of sending the message by the client to the server is decreased; when the comparison result is that the delay is greater than or equal to the first threshold and less than the second threshold, maintaining the client The frequency at which the message is sent to the server; when the comparison result is that the delay is less than the first threshold, the frequency at which the client sends a message to the server is maintained or the frequency at which the client sends a message to the server is increased.

The present invention also provides another optional embodiment, when the comparison result is that the delay is greater than or equal to the second When the threshold is determined, it is determined whether the first N-1 comparison results before the current comparison result are all the time delays greater than the second threshold, that is, whether there are already N comparison results, the delay is greater than or equal to the second threshold, and the current comparison result That is, the result of the Nth comparison; when the judgment result is YES, the frequency at which the client sends a message to the server is decreased, and when the judgment result is no, the frequency at which the client sends a message to the server is maintained; If the delay is less than the first threshold, it is determined whether the first N-1 comparison results before the current comparison result are all less than the first threshold, that is, whether there is already N comparison results, the delay is less than the second threshold. Similarly, the current comparison result is the Nth comparison result; when the judgment result is yes, the frequency of the message sent by the client to the server is increased, and when the judgment result is no, the frequency at which the client sends the message to the server is maintained; The result is that when the delay is greater than or equal to the first threshold and less than the second threshold, the frequency of sending the message to the server by the client is directly maintained.

In this embodiment, the type of information that the client interacts with the server may include at least one of a REQMODE message, an Option message, or a Cache message.

Embodiment 2:

FIG. 2 is a schematic structural diagram of a network congestion control apparatus according to Embodiment 2 of the present invention. The network congestion control apparatus provided in this embodiment may include: a delay acquisition module 201, a delay comparison module 202, and a policy control module 203.

The delay obtaining module 201 is configured to: obtain a delay for reflecting a network state;

Optionally, as shown in FIG. 3, which is a schematic structural diagram of a delay acquisition module in the network congestion control apparatus provided in FIG. 2, in this embodiment, the delay acquisition module 201 includes a transmission delay acquisition submodule. 2011, processing at least one of a delay acquisition sub-module 2012 and an overall delay acquisition sub-module 2013.

The transmission delay acquisition sub-module 2011 is configured to: obtain a transmission delay, which is a time that the information exchanged between the client and the server in the transmission process, and thus the transmission delay can represent the condition of the transmission network. The transmission delay acquisition sub-module 2011 may include one of the following three structures:

The first structure, as shown in FIG. 4, is a schematic structural diagram of a transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the transmission delay acquisition submodule 2011 shown in FIG. The first obtaining unit 401, the second obtaining unit 402, and the first calculating unit 403.

When the client sends a message to the server, the first obtaining unit 401 is configured to: intercept the sending time of the message as the first time point, and the second obtaining unit 402 is configured to: intercept the time when the server receives the message as the second time. At the time point, the first calculating unit 403 is configured to: calculate a time difference between the second time point acquired by the second obtaining unit 402 and the first time point acquired by the first acquiring unit 401, that is, a transmission delay. The delay between the second time point and the first time point is only a one-way transmission delay. Similarly, the transmission delay acquisition sub-module 2011 can also acquire another one-way transmission delay, which is obtained by the second transmission delay acquisition sub-module 2011.

The second structure, as shown in FIG. 5, is a schematic structural diagram of another transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the transmission delay acquisition submodule 2011 shown in FIG. The third obtaining unit 501, the fourth obtaining unit 502, and the second calculating unit 503 are included. When the server sends a message to the client, the third obtaining unit 501 is configured to: intercept the sending time of the message as the third time point, and the fourth obtaining unit 502 is configured to: intercept the time when the client receives the message as the fourth time. At the time point, the second calculating unit 503 is configured to calculate a time difference between the fourth time point acquired by the fourth obtaining unit 502 and the third time point acquired by the third acquiring unit 501 as the transmission delay.

Optionally, in another possible implementation manner of the invention, the sum of the two unidirectional transmission delays may be obtained as a transmission delay, where the transmission delay belongs to the bidirectional transmission delay by the third transmission delay. Get the submodule 2011 get.

The third structure, as shown in FIG. 6, is a schematic structural diagram of another transmission delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the transmission delay acquisition submodule 2011 shown in FIG. The fifth obtaining unit 601, the sixth obtaining unit 602, the seventh obtaining unit 603, the eighth obtaining unit 604, and the third calculating unit 605, when the client sends a message to the server, the fifth obtaining unit 601 is configured to: intercept The sending time of the message is used as the first time point, and the sixth obtaining unit 602 is configured to: intercept the time when the server receives the message as the second time point, and when the server sends a message to the client, the seventh acquiring unit 603 sets To: intercept the sending time of the message as the third time point, the eighth obtaining unit 604 is configured to: intercept the time when the client receives the message as the fourth time point, and the third calculating unit 605 is configured to: calculate the sixth acquiring The difference between the second time point acquired by the unit 602 and the first time point acquired by the fifth obtaining unit 601 is used as the first transmission delay, and the fourth time point and the seventh acquired by the eighth obtaining unit 604 are calculated. The difference between the third time points acquired by the obtaining unit 603 is used as the second transmission delay, and then the sum of the first transmission delay and the second transmission delay is calculated as the transmission delay.

Optionally, the processing delay acquisition sub-module 2012 is configured to: obtain a processing delay, where the processing delay may be a time taken by the server to process the message sent by the client, that is, the client receives the request message from the server to the service. The time between sending a response message to the client, and the processing delay reflects the processing power of the server. As shown in FIG. 7, a schematic diagram of a processing delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, the processing delay acquisition submodule 2012 shown in FIG. 7 includes: a ninth acquisition The unit 701, the tenth obtaining unit 702, and the fourth calculating unit 703, when acquiring the processing delay, the ninth obtaining unit 701 is configured to: intercept the time when the server receives the request message sent by the client as the second time point. The tenth obtaining unit 702 is configured to: intercept the time when the server sends the response message to the client as the third time point, and the fourth calculating unit 703 is configured to: compare the second time point acquired by the tenth obtaining unit 702 with the ninth acquiring unit. At a third time point acquired by 701, a time difference between the third time point and the second time point is obtained as a processing delay.

Optionally, the overall delay acquisition sub-module 2013 can have multiple structures:

The first structure, as shown in FIG. 8, is a schematic structural diagram of an overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the overall delay acquisition submodule 2013 shown in FIG. The eleventh obtaining unit 801, the twelfth obtaining unit 802, and the fifth calculating unit 803 are configured to: acquire the time when the client sends the request message to the server as the first time point, and the twelfth acquiring unit The 802 is set to: obtain the server to send the response message to the third time point, and the fifth calculating unit 803 is configured to: calculate the third time point acquired by the twelfth acquiring unit 802 and the first acquired by the eleventh acquiring unit 801 The difference in time points as the overall delay;

The second structure, as shown in FIG. 9, is a schematic structural diagram of another overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the overall delay acquisition submodule shown in FIG. The 2013 includes a thirteenth obtaining unit 901, a fourteenth obtaining unit 902, and a sixth calculating unit 903. The thirteenth obtaining unit 901 is configured to: obtain a time when the server receives the request message sent by the client, and the time is the second time point. The fourteenth obtaining unit 902 is configured to: acquire the time when the client receives the response message sent by the server is the fourth time point, and the sixth calculating unit 903 is configured to: calculate the fourth time point acquired by the fourteenth obtaining unit 902 The difference between the second time points acquired by the thirteenth obtaining unit 901 is taken as an overall delay;

The third structure, as shown in FIG. 10, is a schematic structural diagram of another overall delay acquisition submodule in the network congestion control apparatus provided in the embodiment shown in FIG. 3, and the overall delay acquisition submodule shown in FIG. 2013 includes a fifteenth obtaining unit 1001, a sixteenth obtaining unit 1002, and a seventh calculating unit 1003; The fifteenth obtaining unit 1001 is configured to acquire the time when the client sends the request message to the server is the first time point, and the sixteenth obtaining unit 1002 is configured to acquire the time when the client receives the response message sent by the server is the fourth time. The seventh calculation unit is set to 1003 to calculate a difference between the fourth time point acquired by the sixteenth acquisition unit 1002 and the first time point acquired by the fifteenth acquisition unit 1001 as the overall time delay.

The overall delay acquisition sub-module 2013 obtains the overall delay by using the direct acquisition method. Since the overall delay is the sum of the transmission delay and the processing delay, the overall delay acquisition sub-module 2013 can be separately obtained. The transmission delay and the processing delay are calculated, and then the sum of the transmission delay and the processing delay is calculated to obtain the overall delay. Therefore, the overall delay acquisition sub-module 2013 can also be three other structures:

The fourth overall delay acquisition sub-module obtains the time when the client sends the request message to the server as the first time point, and obtains the time when the server receives the request message as the second time point, and the second time point and the first time The difference between the time points is used as the first transmission delay, and the time when the server sends the response message is taken as the third time point, and the difference between the third time point and the second time point is taken as the processing delay; the first transmission delay is calculated. Handling the sum of delays as an overall delay;

The fifth overall delay acquisition sub-module obtains the time when the server sends the response message as the third time point, and obtains the time when the client receives the response message as the fourth time point, and the fourth time point and the third time point The difference is used as the second transmission delay, and the time when the server receives the request message sent by the client is used as the second time point, and the difference between the third time point and the second time point is used as the processing delay; the second transmission is calculated. The sum of the delay and the processing delay as the overall delay;

The sixth overall delay acquisition sub-module obtains the time when the client sends the request message to the server as the first time point, and obtains the time when the server receives the request message as the second time point, and the second time point and the first time The difference between the points is used as the first transmission delay, and the time at which the server sends the response message to the client is taken as the third time point, and the time when the client receives the response message is obtained as the fourth time point, and the fourth time point and the third time point are obtained. The difference between the time points is used as the second transmission delay, and the sum of the first transmission delay and the second transmission delay is calculated as the transmission delay, and the difference between the third time point and the second time point is used as the processing delay, and the calculation is calculated. The sum of the transmission delay and the processing delay is taken as the overall delay.

Optionally, when the delay obtaining module 201 obtains the delay, the time stamp may be set in the information that the client interacts with the server, that is, in the request message or the response message. The extension field is added in one less case, and the party sending the information in the client and the server records the timestamp of sending the request message or the response message in the extension field, and when the other party receives the information, the timestamp included in the message is included. Compare with your own system time and calculate the delay of the message during transmission. For example, in an embodiment of the present invention, the transmission delay acquisition sub-module 2011 may set a timestamp of the sending time in the request message when the client sends the request message to the server, when the server receives the request message. At the same time, the transmission delay acquisition sub-module 2011 obtains a timestamp from the server, and compares with the system time of the server to obtain the transmission delay of the request message.

Optionally, in this embodiment, the party that sets the timestamp is the party that calculates the time delay. For example, when the client sends the request message to the server, the timestamp of the sending time is set in the request message. When the server receives the request message and sends a response message, the server copies the timestamp into the extension field of the response message. When the client receives the response message sent by the server, it obtains the timestamp from the system and the system. By comparing the time, the sum of the transmission delay and the processing delay can be calculated to reflect the overall network condition. This method of obtaining the delay is not required to set the time stamp by one of the client or the server, and the other party does not need to require the system time synchronization between the server and the client.

Optionally, the present invention further provides an embodiment, the delay obtaining module 201 obtains a delay by setting a counter, and can start timing by using a counter when the client sends a request message to the server, when the client receives When the response message from the server stops timing, this method is simple, and there is no need to add an extension field in the interaction information between the client and the server, and no calculation is needed.

The delay comparison module 202 is configured to: compare the delay acquired by the delay acquisition module 201 with a preset delay threshold; optionally, as shown in FIG. 11, the network provided in the embodiment shown in FIG. Hold A structure diagram of a delay comparison module in the plug control device, the delay comparison module 202 includes a first threshold comparison sub-module 2021 and a second threshold comparison sub-module 2022; after the delay acquisition module 201 acquires the delay, the first threshold The comparison sub-module 2021 is configured to compare the delay obtained by the delay acquisition module 201 with the first threshold, and obtain a result. The second threshold comparison sub-module 2022 is configured to: obtain the delay obtained by the delay acquisition module 201. The second threshold is compared, and the result is obtained. When the comparison result of the two comparison sub-modules is that the value of the delay is greater than or equal to the second threshold, the current network state is bad; when the value of the delay is greater than or If the first threshold is less than the second threshold, the current network status is normal. When the value of the delay is less than the first threshold, the current network status is good.

The threshold value for comparison between the first threshold comparison sub-module 2021 and the second threshold comparison sub-module 2022 can be set by the user autonomously, and the user can flexibly set according to the storage space and bandwidth capacity of the network used by the user and the performance of the processor. In practical applications, the value of the first threshold may be equal to the value of the second threshold.

The policy control module 203 is configured to: control the sending policy of the client according to the comparison result obtained by the delay comparison module 202.

After the delay comparison module 202 obtains the comparison result, the policy control module 203 can control the data transmission policy of the current client according to the comparison result. Optionally, when the comparison result is that the delay is greater than or equal to the second threshold, the policy control module 203 is configured to reduce the frequency at which the client sends the message to the server; when the comparison result is that the delay is greater than or equal to the first threshold and is less than the first The second time threshold, the policy control module 203 is configured to maintain the frequency at which the client sends a message to the server. When the comparison result is that the delay is less than the first threshold, the policy control module 203 is configured to maintain the client sending the message to the server. Frequency or increase the frequency with which clients send messages to the server.

The present invention further provides another optional embodiment. As shown in FIG. 12, a schematic structural diagram of a policy control module in the network congestion control apparatus provided in the embodiment shown in FIG. 2, the policy control module 203 includes a judgment. The sub-module 2031 and the control sub-module 2032, on the one hand, the judging sub-module 2031 is set to: when the comparison result obtained by the current-span comparison module 202 is that the delay is greater than the second threshold, the first N-1 comparison results before the current comparison result are determined. Whether the delay is greater than the second threshold, that is, determining whether there are already N comparison results, the delay is greater than the second threshold, and the current comparison result is the Nth comparison result; the control sub-module 2032 is set to: when determining When the determination result of the submodule 2031 is YES, the frequency at which the client sends a message to the server is decreased, and when the judgment result of the submodule 2031 is negative, The frequency of the message sent by the client to the server is maintained. On the other hand, the determining sub-module 2031 is further configured to: when the comparison result obtained by the time delay comparison module 202 is that the delay is less than the first threshold, before the current comparison result is determined Whether the N-1 comparison result is less than the first threshold, that is, whether there is already N comparison results, the delay is less than the second threshold, and the current comparison result is also the Nth comparison result; accordingly, the controller The module 2032 is further configured to: when the judgment result of the determining submodule 2031 is YES, increase the frequency at which the client sends a message to the server, and when the judgment result of the submodule 2031 is negative, the client is sent to send a message to the server. On the other hand, the control sub-module 2032 is further configured to: when the comparison result obtained by the delay comparison module 202 is that the delay is greater than or equal to the first threshold and less than the second threshold, directly maintain the frequency at which the client sends a message to the server. .

Optionally, in another embodiment provided by the present invention, as shown in FIG. 13, a schematic structural diagram of another policy control module in the network congestion control apparatus provided in the embodiment shown in FIG. 2, a control policy adjustment module In addition to the determining sub-module 2031 and the control sub-module 2032, the 203 may further include a policy storage sub-module 2033, configured to: store a control policy adjustment record that has been performed in a certain period, where the day is used as the above-mentioned period, because in one day The interaction between the client and the server has a busy time and a free time. In the long-term interaction, the busy hour and the idle time of each day are in a relatively fixed time period, so when the policy storage sub-module 2033 stores one day. After the control policy adjustment record is made, the control sub-module 2032 can adjust the control policy according to the record provided by the policy storage sub-module 2033, so that the adjustment of the control strategy of the present invention is before the network condition, and the defect of the hysteresis adjustment is avoided. .

Embodiment 3:

On the basis of the second embodiment, the embodiment of the present invention further provides a client, which includes the network congestion control device provided in any of the embodiments shown in FIG. 2 to FIG.

The client provided in the embodiment of the present invention may be a client supporting an Internet Content Adaptation Protocol (ICAP). For example, the client may be a personal computer or a terminal that supports the ICAP protocol software. It can also be a carrier-class network node device that implements the ICAP protocol. The client obtains related resources for service processing by sending a related request message to the server. Correspondingly, the server used in conjunction with the client provided by the embodiment of the present invention may be a personal computer, a terminal or a commercial server that supports the ICAP protocol software, or a carrier-class network node device that implements the ICAP protocol.

One of ordinary skill in the art will appreciate that all or a portion of the steps of the above-described embodiments can be implemented using a computer program flow, which can be stored in a computer readable storage medium, such as on a corresponding hardware platform (eg, The system, device, device, device, etc. are executed, and when executed, include one or a combination of the steps of the method embodiments.

Alternatively, all or part of the steps of the above embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve.

The devices/function modules/functional units in the above embodiments may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.

When the device/function module/functional unit in the above embodiment is implemented in the form of a software function module and sold or used as a stand-alone product, it can be stored in a computer readable storage medium. The above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Industrial applicability

In the embodiment of the present invention, the time delay of the network state is obtained, and the obtained delay is compared with the preset delay threshold to learn the current network situation, and then the policy of the client is controlled according to the comparison result, and the real-time is achieved. Understand the current network status and adjust the data transmission policy of the client according to the current network status. The related technology solves the problem that the client does not understand the network environment or the processing capability of the server, and always interacts with a single data transmission policy. The problem that causes the network environment to deteriorate is improved, and the efficiency of the interaction between the client and the server is improved, and the network resource configuration is optimized.

Claims

A network congestion control method includes:

Obtain a delay for reflecting the state of the network;

Comparing the delay with a preset delay threshold;

Control the client's sending policy based on the comparison result.
The network congestion control method according to claim 1, wherein the acquired delay is a transmission delay that characterizes a transmission network environment, a processing delay that characterizes a processing capability of the server, or an overall delay that characterizes the network environment.
The network congestion control method according to claim 2, wherein when the obtained delay is the transmission delay, the manner of obtaining the transmission delay includes one of the following three types:

Obtaining, by the client, a time for sending the first message to the server as a first time point, acquiring a time when the server receives the first message as a second time point, and using the second time point The difference between the first time points is used as the transmission delay;

Obtaining, by the server, a time for sending the second message to the client as a third time point, acquiring a time when the client receives the second message as a fourth time point, and using the fourth time point The difference between the third time points is used as the transmission delay;

Obtaining, by the client, a time for sending the first message to the server as a first time point, acquiring a time when the server receives the first message as a second time point, and using the second time point The difference between the first time point is used as the first transmission delay; the time when the server sends the second message to the client is used as the third time point, and the client receives the second message. The time is taken as the fourth time point, and the difference between the fourth time point and the third time point is used as the second transmission delay, and the sum of the first transmission delay and the second transmission delay is calculated as The transmission delay.
The network congestion control method according to claim 2, wherein when the obtained delay is the processing delay, the manner of obtaining the processing delay includes:

Obtaining, as the second time point, the time when the server receives the first message sent by the client, and acquiring the time when the server sends the second message to the client, as the third time point, where the The difference between the three time points and the second time point is taken as the processing delay.
The network congestion control method according to claim 2, wherein when the obtained delay is the overall time delay, the manner of obtaining the overall delay includes one of the following three types:

Obtaining, as the first time, the time when the client sends the first message to the server, and acquiring the time when the server sends the second message to the client as the third time point, where the third time is a difference between the point and the first time point as the overall time delay;

Obtaining, as the second time point, the time when the server receives the first message sent by the client, and acquiring the time when the client receives the second message sent by the server, as the fourth time point, Determining a difference between the fourth time point and the second time point as the overall time delay;

Obtaining, by the client, a time for sending the first message to the server as a first time point, and acquiring a time when the client receives the second message sent by the server as a fourth time point, where the The difference between the four time points and the first time point is taken as the overall time delay.
The network congestion control method according to any one of claims 1 to 5, wherein the preset delay threshold includes a first threshold and a second threshold, and the first threshold is smaller than the second threshold.
The network congestion control method according to claim 6, wherein the controlling the sending policy of the client according to the comparison result comprises:

When the comparison result is that the delay is greater than or equal to the second threshold, reducing a frequency at which the client sends a message to the server;

And maintaining, when the comparison result is that the delay is greater than or equal to the first threshold and less than the second threshold, maintaining a frequency at which the client sends a message to the server;

And when the comparison result is that the delay is less than the first threshold, maintaining a frequency at which the client sends a message to the server or increasing a frequency at which the client sends a message to the server.
The network congestion control method according to claim 6, wherein the controlling the sending policy of the client according to the comparison result comprises:

When the comparison result is that the delay is greater than or equal to the second threshold, determining whether the first N-1 comparison results before the current comparison result are all the delay is greater than or equal to the second threshold; If the result of the determination is yes, the frequency at which the client sends a message to the server is decreased, and when the determination result is no, the frequency at which the client sends a message to the server is maintained;

When the comparison result is that the delay is greater than or equal to the first threshold and less than the second gate Limiting the frequency at which the client sends a message to the server;

When the comparison result is that the delay is less than the first threshold, determining whether the first N-1 comparison results before the current comparison result are all the delay is less than the first threshold; when the judgment result is yes The frequency of the message sent by the client to the server is increased, and when the determination result is no, the frequency at which the client sends a message to the server is maintained.
A network congestion control device includes:

The delay acquisition module is configured to: obtain a delay for reflecting a network state;

The delay comparison module is configured to: compare the delay acquired by the delay acquisition module with a preset delay threshold;

The policy control module is configured to: control a sending policy of the client according to the comparison result obtained by the delay comparison module.
The network congestion control apparatus according to claim 9, wherein the delay acquisition module comprises at least one of a transmission delay acquisition sub-module, a processing delay acquisition sub-module, and an overall delay acquisition sub-module;

The transmission delay acquisition submodule is configured to: acquire a transmission delay that characterizes a transmission network environment;

The processing delay acquisition submodule is configured to: acquire a processing delay that represents a processing capability of the server;

The overall delay acquisition sub-module is configured to: obtain an overall delay that characterizes the network environment.
The network congestion control apparatus according to claim 10, wherein when the delay acquisition module includes the transmission delay acquisition submodule, the transmission delay acquisition submodule comprises:

The first obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a second acquiring unit, configured to: acquire a time when the server receives the first message as a second time point;

a first calculating unit, configured to: calculate, as the transmission delay, a difference between the second time point acquired by the second acquiring unit and the first time point acquired by the first acquiring unit;

Alternatively, the transmission delay acquisition submodule includes:

a third obtaining unit, configured to: obtain, by the server, a second message sent to the client Time as the third time point;

a fourth obtaining unit, configured to: acquire a time when the client receives the second message as a fourth time point;

a second calculating unit, configured to: calculate, as the transmission delay, a difference between the fourth time point acquired by the fourth acquiring unit and the third time point acquired by the third acquiring unit;

Alternatively, the transmission delay acquisition submodule includes:

The fifth obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a sixth obtaining unit, configured to: acquire a time when the server receives the first message as a second time point;

The seventh obtaining unit is configured to: acquire a time when the server sends the second message to the client as a third time point;

An eighth obtaining unit, configured to: acquire a time when the client receives the second message as a fourth time point;

a third calculating unit, configured to: calculate a difference between the second time point acquired by the sixth acquiring unit and the first time point acquired by the fifth acquiring unit as a first transmission delay, and calculate the a difference between the fourth time point acquired by the eighth obtaining unit and the third time point acquired by the seventh acquiring unit as a second transmission delay, and calculating the first transmission delay and the second The sum of transmission delays is taken as the transmission delay.
The network congestion control apparatus according to claim 10, wherein when the delay acquisition module includes the processing delay acquisition sub-module, the processing delay acquisition sub-module comprises:

The ninth obtaining unit is configured to: acquire, as the second time point, the time when the server receives the first message sent by the client;

The tenth acquiring unit is configured to: acquire, as the third time point, a time when the server sends the second message to the client;

The fourth calculating unit is configured to: calculate, as the processing delay, a difference between the third time point acquired by the tenth acquiring unit and the second time point acquired by the ninth acquiring unit.
The network congestion control apparatus according to claim 10, wherein when the delay acquisition module includes the overall delay acquisition submodule, the overall delay acquisition submodule comprises:

The eleventh acquiring unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a twelfth obtaining unit, configured to: acquire a time when the server sends a second message to the client as a third time point;

a fifth calculating unit, configured to: calculate, as the overall delay, a difference between the third time point acquired by the twelfth acquiring unit and the first time point acquired by the eleventh acquiring unit;

Alternatively, the overall delay acquisition submodule includes:

a thirteenth obtaining unit, configured to: acquire a time when the server receives the first message sent by the client as a second time point;

a fourteenth obtaining unit, configured to: acquire a time when the client receives the second message sent by the server as a fourth time point;

a sixth calculating unit, configured to: calculate, as the overall delay, a difference between the fourth time point acquired by the fourteenth acquiring unit and the second time point acquired by the thirteenth acquiring unit;

Alternatively, the overall delay acquisition submodule includes:

The fifteenth obtaining unit is configured to: acquire a time when the client sends the first message to the server as the first time point;

a sixteenth obtaining unit, configured to: the time when the client receives the second message sent by the server is used as a fourth time point;

The seventh calculating unit is configured to: calculate, as the overall delay, a difference between the fourth time point acquired by the sixteenth acquiring unit and the first time point acquired by the fifteenth acquiring unit.
The network congestion control apparatus according to any one of claims 9-13, wherein the delay comparison module comprises a first threshold comparison sub-module and a second threshold comparison sub-module;

The first threshold comparison submodule is configured to: compare the time delay acquired by the delay acquisition module with a first threshold;

The second threshold comparison submodule is configured to: the time delay acquired by the delay acquisition module Compare with the second threshold.
The network congestion control apparatus according to claim 14, wherein the policy control module controls the sending policy of the client according to the comparison result obtained by the delay comparison module, including:

When the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the second threshold, the policy control module is configured to: reduce a frequency at which the client sends a message to the server;

When the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the first threshold and less than the second threshold, the policy control module is configured to: maintain the client to the The frequency at which the server sends messages;

When the comparison result obtained by the delay comparison module is that the delay is less than the first threshold, the policy control module is configured to: maintain a frequency or improve the frequency at which the client sends a message to the server. The frequency at which the client sends a message to the server.
The network congestion control apparatus according to claim 14, wherein the policy control module further comprises a judgment submodule and a control submodule:

The determining sub-module is configured to: when the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the second threshold, determine whether the first N-1 comparison results before the current comparison result are all The control sub-module is configured to: when the judgment result of the determining sub-module is YES, reduce the frequency at which the client sends a message to the server, when And when the determining result of the determining submodule is negative, maintaining a frequency at which the client sends a message to the server;

The control sub-module is further configured to: when the comparison result obtained by the delay comparison module is that the delay is greater than or equal to the first threshold and less than the second threshold, maintaining the client to the service The frequency at which the message is sent by the terminal;

The determining sub-module is further configured to: when the comparison result obtained by the delay comparison module is that the delay is less than the first threshold, determine whether the first N-1 comparison results before the current comparison result are all The control sub-module is further configured to: when the determination result of the determining sub-module is YES, increase the frequency at which the client sends a message to the server, when When the judgment result of the submodule is negative, the client is sent to send a message to the server. frequency.
A client comprising the network congestion control device of any of claims 9-16.