WO2014153780A1 - Network congestion processing method, network node, and network system - Google Patents

Abstract

Embodiments of the present invention provide a network congestion processing method, a network node, and a network system. The method comprises: an ingress node of a network receiving network congestion information sent by an egress node, the network congestion information comprising an address of the egress node, a congestion amount, and an address of an intermediate node where congestion occurs; determining a path of network congestion according to the address of the egress node and the address of the intermediate node; determining, according to the path, a data stream that needs to pass through the path and is in the data stream entering a network; and controlling, according to the congestion amount, user subscription information, and a network control policy, the data stream that needs to pass through the path. Technical solutions of the embodiments of the present invention can effectively process the network congestion, and improve a congestion processing effect.

Description

 The present invention relates to a network communication technology, and in particular, to a network congestion processing method, a network node, and a network system. Background technique

 Due to the limited data transmission capacity of the network, when the data traffic transmitted in the network exceeds the transmission capacity of the network, congestion will occur in the network. For example, when the intermediate nodes in the path network, such as the data traffic of the router exceeds the processing capacity of the router, When the network congestion occurs on the router, network congestion needs to be processed at this time, otherwise the data transmission of the normal network will be affected. At present, when a network node in a network is congested, it is usually processed by dropping data packets to prevent network congestion from affecting the entire network data transmission. However, the method of dropping the data packet will be corresponding. The business has a negative impact. For example, for the data transmission of VoIP and video services, if the method of dropping the data packet is used, the user's use effect of these services will be affected. Meanwhile, when the data packet is discarded, the data is discarded. After the data packet is discarded, the sender still retransmits the data packet, which causes waste of resources and also increases congestion.

An explicit congestion explicit notification (ECN) mechanism for performing congestion control is also proposed in the prior art, which is an intermediate node in the network by deploying an ECN mechanism at the transmitting end of the data and the receiving end of the data. When congestion occurs, the receiving end notifies the sending end to control the data sent to the receiving end. Since the data is controlled by the transmitting end of the data, the entire network is not considered, resulting in poor congestion processing, and When the ECN mechanism is not supported at the receiving end and the transmitting end, congestion processing cannot be performed. To this end, the prior art also proposes a network-based congestion processing mechanism, namely a Pre-Congestion Notification (PCN) mechanism, which deploys a PCN mechanism at the ingress node and the egress node of the network, and the intermediate node Congestion marking the data packets of the inelastic data stream to forward the marked inelastic data packet as a congested data packet, and the egress node obtains the current network congestion by counting the number of congested data packets, and sends the network congestion condition. To the ingress node, so that the ingress node can enter the network through the ingress node based on the congestion condition The traffic of the inelastic data stream is controlled to reduce network congestion.

 However, in the existing network congestion processing using the PCN mechanism, after receiving the congestion notification, the ingress node limits all inelastic data entering the network through the ingress node, and part of the inelastic data entering the network from the ingress node is often not Congestion data, which causes poor network congestion control and poor congestion processing. At the same time, existing ingress nodes only restrict inelastic data streams, but network congestion on elastic data streams cannot be performed. deal with. Summary of the invention

 The embodiment of the invention provides a network congestion processing method, a network node and a network system, which can effectively improve the pertinence of network congestion processing data and improve the congestion processing effect.

 The first aspect of the present invention provides a network congestion processing method, including: an ingress node of a network receives network congestion information sent by an egress node, where the network congestion information includes an address, a congestion, and congestion of the egress node. The address of the intermediate node;

 Determining a path of network congestion according to an address of the egress node and an address of the intermediate node;

 Determining, according to the path, a data stream that needs to pass through the path in a data stream entering the network;

 And controlling the data flow that needs to pass through the path according to the congestion amount, the user subscription information, and the network control policy.

 With reference to the first aspect, in a first possible implementation, the controlling the data flow that needs to go through the path according to the congestion quantity, the user subscription information, and the network control policy includes:

 Determining, according to the user subscription information, a service corresponding to each data flow that needs to go through the path, and determining, according to the network control policy, a priority of a service corresponding to each data flow that needs to pass the path;

 In each data stream that needs to pass through the path, the data stream corresponding to the service with a lower priority is preferentially controlled according to the congestion amount.

With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, the determining, according to the network control policy, a priority of a service corresponding to each data flow that needs to pass the path, includes: Determining the priority of the service corresponding to the data flow passing the path according to the priority of the geographic location of the path set in the network control policy.

 With reference to the first aspect or the first or the second possible implementation manners of the first aspect, in a third possible implementation manner, the determining, according to the congestion amount, the user subscription information, and the network control policy, The data flow is controlled, including:

 According to the congestion amount, the transmission rate of all data flows that need to pass through the path is controlled, or the transmission rate of at least a part of the data streams is restricted or prohibited from being transmitted.

 In a second aspect, an embodiment of the present invention provides a network congestion processing method, including: receiving, by an egress node of a network, a congested data packet sent by an intermediate node, where the congested data packet is used to forward a forwarded data packet when the intermediate node generates congestion. Tagged packet;

 Determining the amount of congestion according to the number of received congested data packets;

 Sending network congestion information to an ingress node of the network, where the network congestion information includes: the congestion amount, an address of the egress node, and an address of the intermediate node, so that the ingress node determines based on the network congestion information A path of network congestion and control of data flows that enter the network through the ingress node and pass through the path.

 In a third aspect, an embodiment of the present invention provides a network node, including:

 a receiving module, configured to receive network congestion information sent by an egress node of the network, where the network congestion information includes an address of the egress node, a congestion amount, and an address of an intermediate node where congestion occurs;

 a processing module, configured to determine a path of network congestion according to an address of an egress node and an address of an intermediate node in the network congestion information received by the receiving module, and determine, according to the path, a data flow entering the network And a data flow of the path, and controlling, according to the congestion amount, user subscription information, and a network control policy, the data flow that needs to pass through the path.

 With reference to the third aspect, in a first possible implementation, the processing module is specifically configured to determine, according to the user subscription information, a service corresponding to each data flow that needs to pass the path, and determine, according to the network control policy, The priority of the service corresponding to each data flow of the path is required, and in each data flow that needs to pass through the path, the data flow corresponding to the service with a lower priority is preferentially controlled according to the congestion amount.

In combination with the first possible implementation of the third aspect, in the second possible implementation, The processing module is specifically configured to determine a priority of a service corresponding to the data flow passing the path according to a priority of a geographic location of the path set in the network control policy.

 With reference to the third aspect or the first or the second possible implementation manner of the third aspect, in a third possible implementation manner, the processing module is specifically configured to: according to the congestion quantity, all data that needs to pass the path The transmission rate of the stream is controlled, or the transmission rate of at least a part of the data stream is restricted or prohibited from being transmitted.

 In a fourth aspect, an embodiment of the present invention provides a network node, including a receiver, a memory, and a processor connected to the memory, where the memory stores a set of program codes, where: the receiver is configured to receive a network. Network congestion information sent by the egress node, where the network congestion information includes an address of the egress node, a congestion amount, and an address of an intermediate node where congestion occurs;

 The processor is configured to execute program code stored in the memory, to determine a path of network congestion according to an address of an egress node and an address of an intermediate node in the network congestion information received by the receiver, and The path determines a data flow that needs to pass through the path in the data flow entering the network, and controls the data flow that needs to pass through the path according to the congestion amount, the user subscription information, and the network control policy.

 With reference to the fourth aspect, in a first possible implementation manner, the processor is specifically configured to determine, according to the user subscription information, a service corresponding to each data flow that needs to pass the path, and determine, according to the network control policy, The priority of the service corresponding to each data flow of the path is required, and in each data flow that needs to pass through the path, the data flow corresponding to the service with a lower priority is preferentially controlled according to the congestion amount.

 With reference to the first possible implementation manner of the fourth aspect, in a second possible implementation manner, the processor is specifically configured to determine, according to a priority of a geographic location of a path set in the network control policy, The priority of the service corresponding to the data flow of the path.

 With reference to the fourth aspect or the first or the second possible implementation manner of the fourth aspect, in a third possible implementation, the processor is specifically configured to: according to the congestion quantity, all data that needs to pass the path The transmission rate of the stream is controlled, or the transmission rate of at least a part of the data stream is restricted or prohibited from being transmitted.

 In a fifth aspect, an embodiment of the present invention provides a network node, including:

a receiving module, configured to receive a congestion data packet sent by the intermediate node, where the congestion data packet is The intermediate node generates a data packet that marks the forwarded data packet when the congestion occurs; the processing module is configured to determine a congestion amount according to the quantity of the congestion data packet received by the receiving module, and obtain network congestion information, where The network congestion information includes: the congestion amount, an address of the egress node, and an address of the intermediate node;

 a sending module, configured to send the network congestion information obtained by the processing module to an ingress node of the network, so that the ingress node determines a path of network congestion based on the network congestion information and enters a network through the ingress node And control through the data flow of the path.

 In a sixth aspect, an embodiment of the present invention provides a network node, including a receiver, a transmitter, a memory, and a processor connected to the memory, where the memory stores a set of program code, where:

 The receiver is configured to receive a congestion data packet sent by an intermediate node, where the congestion data packet is a data packet that marks the forwarded data packet when the intermediate node generates congestion;

 The processor is configured to execute program code stored in the memory to determine a congestion amount according to the number of the congestion data packets received by the receiver, and obtain network congestion information, where the network congestion information includes : the amount of congestion, an address of the egress node, and an address of the intermediate node;

 Transmitting, by the transmitter, the network congestion information obtained by the processor to an ingress node of the network, so that the ingress node determines a path of network congestion based on the network congestion information and passes the ingress node Enter the network and control the data flow through the path.

 According to a seventh aspect, an embodiment of the present invention provides a network system, including: an ingress node, an intermediate node, and an egress node, where the ingress node is the third or third or third possibility of using the third aspect or the third aspect The network node provided by the implementation manner or the first or second or three possible implementation manners of the fourth aspect, where the egress node is the network node provided by the fifth aspect or the sixth aspect.

The embodiment of the present invention can determine the path of the network congestion by using the egress node carried in the network congestion information and the intermediate node causing the congestion, so as to control the data flow of the path that needs to go through the congestion in the network, so as to reduce the path through the path. The data traffic makes the network congestion processing data more targeted, which can effectively provide network congestion processing effects and reduce or eliminate network congestion. DRAWINGS 1 is a schematic flowchart of a network congestion processing method according to Embodiment 1 of the present invention; FIG. 2 is a schematic flowchart of a network congestion processing method according to Embodiment 2 of the present invention; FIG. 2B is a schematic diagram of a network scenario according to Embodiment 2 of the present invention;

 3A is a schematic diagram of a tunnel network scenario of a network congestion processing method applied to a network using a tunneling technology according to Embodiment 3 of the present invention;

 FIG. 3B is a schematic flowchart of a network congestion processing method according to Embodiment 3 of the present invention; FIG. 4 is a schematic flowchart of a network congestion processing method according to Embodiment 4 of the present invention; FIG. Schematic diagram

 6 is a schematic structural diagram of a network node according to Embodiment 6 of the present invention;

 7 is a schematic structural diagram of a network node according to Embodiment 7 of the present invention;

 8 is a schematic structural diagram of a network node according to Embodiment 8 of the present invention;

 FIG. 9 is a schematic structural diagram of a network system according to Embodiment 9 of the present invention.

detailed description

 FIG. 1 is a schematic flowchart diagram of a network congestion processing method according to Embodiment 1 of the present invention. The embodiment can be applied to the network-based congestion processing, and the data of the network congestion can be controlled at the ingress node of the network to reduce or eliminate the network congestion. Specifically, as shown in FIG. Including the following steps:

 Step 101: The ingress node of the network receives network congestion information sent by the egress node, where the network congestion information includes an address of the egress node, a congestion amount, and an address of an intermediate node where congestion occurs;

 Step 102: The ingress node determines a path of network congestion according to an address of the egress node and an address of the intermediate node.

 Step 103: The ingress node determines, according to the path, a data flow that needs to go through the path in the data flow that enters the network.

 Step 104: The ingress node controls, according to the congestion amount, the user subscription information, and the network control policy, the data flow that needs to pass the path.

In this embodiment, the egress node of the network may collect and obtain the congestion of the intermediate node of the network, obtain the congestion information, and send the congestion information to the ingress node, so that the ingress node can enter the network based on the foregoing steps 101-step 104. And control the data flow through the congested path Reduces data traffic on the congestion path in the network, reducing or eliminating network congestion.

 It can be understood by those skilled in the art that, in the foregoing step 103, the data flow that determines the path that enters the network and passes the determined network congestion may be one or more data flows, and controls the data flow, that is, from the ingress node and through the network. The data flow of the congested path is controlled.

 In summary, the network congestion processing method provided by the embodiment of the present invention can determine the network congestion path through the egress node carried in the network congestion information and the intermediate node causing the congestion, so that the data flow of the path entering the congestion path can be entered into the network. Control is performed to reduce the data traffic passing through the path, so that the network congestion processing data is highly targeted, effectively providing network congestion processing effects, and reducing or eliminating network congestion.

 2A is a schematic flowchart of a network congestion processing method according to Embodiment 2 of the present invention; and FIG. 2B is a schematic diagram of a network scenario applied according to Embodiment 2 of the present invention. As shown in FIG. 2B, the network applied by the method in this embodiment includes an ingress node, an intermediate node, and an egress node, where the sending end of the data can send data to the network through the ingress node, and the intermediate node can send data in the network to the data. The receiving end, the ingress node, the intermediate node, and the egress node may be various network devices, such as routers, and the like, and one intermediate node may be connected to multiple ingress nodes and multiple egress nodes; A congestion processing node and a congestion information collection node, wherein the network congestion processing node is configured to process network congestion information collected by the congestion information collection node. Specifically, as shown in FIG. 2A and FIG. 2B, the network congestion processing method in this embodiment may include the following steps: Step 201: When the intermediate node detects that the network itself is congested, the intermediate node marks the data packet to be forwarded to obtain a congestion data packet. And forward it.

 Step 202: The egress node receives the data packet forwarded by the intermediate node, and when the data packet is congested, notifies the congestion information collecting node, and the congestion information collecting node counts the number of the congested data packets forwarded by the intermediate node to the egress node. Determine the amount of congestion.

 Step 203: The congestion information collecting node uses the collected congestion amount, the address of the egress node, and the address of the intermediate node as network congestion information, and sends the network congestion information to the network congestion processing node.

Step 204: After receiving the network congestion information, the network congestion processing node determines a path of the network congestion according to the address of the egress node and the address of the intermediate node in the network congestion information, and sends the congested path and the congestion quantity to the ingress node. Step 205: After receiving the congestion path and the congestion amount, the ingress node may control, according to the congestion quantity, a data flow that needs to enter the network from the ingress node and transit the intermediate node and the egress node.

 In the network to which the method of the embodiment is applied, the network congestion processing node and the congestion information collection node may be deployed in the traditional network system to obtain network congestion information and process network congestion information, where the ingress node and the middle are described in FIG. 2B. The node and the egress node are network nodes deployed in the traditional network, the network congestion processing node may be deployed at the ingress node of the network, the congestion information gathering node may be deployed at the egress node of the network, and the congestion information gathering node may be in the network Each ingress node is communicatively coupled, and the congestion information collection node can be in communication with each egress node in the network.

 In the foregoing step 201, the intermediate node may perform congestion marking on the data packet and forward the congestion packet after the congestion flag as long as the network congestion is detected.

 In this embodiment, the data is congested, and the ECN mechanism may be used to perform congestion marking. Specifically, the IP packet may be ECN-marked by using two bits in the service type TOS field in the IP header of the data IP packet. To indicate that the IP packet is a congested packet. The ECN mechanism is used to perform the congestion labeling, which may also be called the ECN label. The specific labeling method is the same as or similar to the existing ECN label, and will not be described here. Those skilled in the art can understand that when the ECN mechanism is used for congestion marking, the ECN mechanism needs to be deployed in each network node in the network, that is, each network node in the network supports the ECN mechanism.

 Those skilled in the art can understand that in practice, in addition to using the ECN mechanism for congestion tagging, other methods may be used, for example, a specific bit in the IP packet is used as a specific tag bit of the congestion packet, by modifying the specific bit. The value can be used to determine if a congestion flag has been made to determine if it is a congested packet.

 It can be understood by those skilled in the art that when the intermediate node in the foregoing step 201 detects that the network is congested by itself, the intermediate node may determine that the network traffic is congested when detecting that the data traffic exceeds a certain range or threshold. For example, the active queue management AQM can be used to detect whether the network is congested. The specific implementation of the network congestion detection is not particularly limited in this embodiment.

In the foregoing step 202, when the egress node receives the congested data packet forwarded by the intermediate node, the egress node can notify the congestion information collecting node to count the number of the congested data packet, because the congested data packet The congestion occurs when the intermediate node is congested. Therefore, the number of received congestion packets can be fed back to the congestion level. Therefore, the congestion amount can be determined according to the number of the congestion data packets, and the congestion amount may specifically refer to congestion. The number of packets. Generally speaking, the more congestion packets indicate the more serious the congestion, the actual application can also indicate the degree of congestion by setting the congestion amount as a percentage value. Specifically, the congestion packet can be counted by the statistical unit time. The number, the percentage value corresponding to the amount of congestion.

 In the foregoing step 203, the congestion information collecting node can specifically send the congestion information when the congestion amount measured in the unit time reaches a preset value, that is, the network congestion processing node is notified only when the congestion amount reaches a certain level. At this time, the congestion amount and the address of the intermediate node and the egress node can be fed back to the network congestion processing node in real time as network congestion information.

 In actual application, the foregoing step 203 may periodically feed back the statistical congestion information to the network congestion processing node, that is, the congestion information collection node may periodically count the number of congestion data packets, and periodically send the information. The network congestion information is sent to the network congestion processing node; or the congestion information collection node may also feed back network congestion information to the network congestion processing node according to the active request of the network congestion processing node, for example, the network congestion processing node detects frequent repeated transmission of the ingress node. When the same data indicates that congestion may occur in the network, the network congestion information may be actively obtained from the congestion information collection node, or the network congestion processing node may actively obtain the current network from the congestion information collection node by sending a command. Congestion information. At this time, the network congestion information may further include information such as the time of counting the congestion amount.

 In the foregoing steps 204 and 205, after receiving the network congestion information, the network congestion processing node may determine the path of the network congestion according to the address of the egress node and the address of the intermediate node in the network congestion information, and may determine the path and the amount of congestion. Feedback to the ingress node, the ingress node controls the data flow that needs to enter the network from the ingress node and route the path to reduce data traffic through the path, reduce or eliminate path congestion, and thereby reduce congestion of the intermediate node.

In the foregoing embodiments of the present invention, the ingress node controls, according to the congestion amount, the user subscription information, and the network control policy, the data flow of the path that needs to be queried by the network, and the method may include: determining, according to the user subscription information, each path that needs to go through the path. a service corresponding to the data flow, and determining, according to the network control policy, a priority of the service corresponding to each data flow that needs to pass through the path; in each data flow of the path that needs to pass through the network congestion, priority is given to the priority according to the congestion amount The data flow corresponding to the low service is controlled. Specifically, the ingress node determines, according to the information such as the network control policy and the user subscription data, the priority of the data corresponding to the service that enters the network through the ingress node and determines the congestion path, and controls the data flow according to the priority. . For example, the service corresponding to each data stream may be determined according to the user subscription data, and the control policy may determine the priority of each service, so that the flow control of the data flow may be performed according to the determined priority of the data flow, and the specific implementation thereof The same or similar to the flow control process of the prior art, and details are not described herein again.

 In the foregoing embodiments of the present invention, the ingress node controls, according to the congestion quantity, the user subscription information, and the network control policy, the data flow of the path that needs to pass through the network congestion, and specifically, all the data flows of the path that needs to pass the congestion according to the congestion amount. The transmission rate is controlled, or the transmission rate of at least a portion of the data stream is limited or prohibited.

 In the above embodiments of the present invention, when the data stream is controlled, the transmission rate may be reduced, for example, by a certain percentage, or by the transmission of some data streams or all data streams.

 It can be understood by those skilled in the art that in the embodiment shown in FIG. 2A and FIG. 2B, the congestion information gathering node can also be integrated with the egress node, and the network congestion processing node can also be integrated with the ingress node, so that The network congestion processing node is integrated at the ingress node of the traditional network, and the congestion information collection node is integrated at the egress node, so that the function of the network congestion processing node is integrated on the existing ingress node, and the congestion information is integrated at the egress node. The function of the node. The specific implementation of the network congestion processing is similar to the implementation process shown in FIG. 2A, and details are not described herein again.

 It can be understood by those skilled in the art that, when the ingress node controls the data flow that needs to pass through the congested path in the data flow that is accessed from the ingress node, the ingress node may determine to enter the network through the ingress node according to a preset configuration or a negotiation manner. The data flow in the data flow needs to go through the congested path, and the specific implementation process will not be described here.

 In the foregoing embodiments of the present invention, the address of the intermediate node and the address of the egress node are specifically IP addresses, so that the congestion path can be determined based on the IP address, and it is also determined that the egress node enters the network and passes the congestion. The data flow of the path.

It can be understood by those skilled in the art that the ingress node, the egress node, and the intermediate node described in the foregoing embodiments of the present invention are only the division of nodes through which data flows in the network, and data in a data stream. During the transmission, a node may be an ingress node or an egress node. In the foregoing embodiments of the present invention, when obtaining the priority of the service corresponding to the data flow of the path that needs to pass through the network congestion, the path of the network congestion may also be determined according to the priority of the geographic location of the path set in the network control policy. The priority of the traffic corresponding to the data stream. After acquiring the path of the network congestion, the geographic location of the path that the network can be congested determines the priority of the service corresponding to the data flow of the path that is congested by the network, so as to control the data flow, that is, according to the egress node and the intermediate node. When the address determines the path of network congestion, it can also obtain the location information of the path of the network congestion, and control the data entering the network through the ingress node in a more accurate and more flexible manner. Specifically, when the ingress node needs to control the data flow of the multiple paths, the priority of the data flow to be controlled may be determined according to the geographical location of each path, so as to determine the data flow control of each path according to the priority. For example, the intermediate node R1 is congested, and the congestion affects the flow of each incoming data from the ingress node to the egress node 1, the egress node 2, and the egress node 3, so that the ingress node needs to go through the egress node. The path 1 of 1 is controlled by the data path of the path 21 of the egress node 2 and the path 3 of the egress node 3, and at this time, the geographic locations of the path 1, the path 2, and the path 3 are acquired, so that the geography can be obtained according to each geography. The importance of the service at the location, on the basis of ensuring important services, preferentially control the data flow of the path through which the non-essential service passes. For example, the location of the egress node 1 is the geographic location 1, and the location of the egress node 2 is the geographic location 2 The location of the exit node 3 is the geographic location 3, while the geographic location 1 and the geographic location 2 are very important, and the geographic location 3 The service is relatively non-essential. At this time, the ingress node can preferentially control the data flow of the path 3 by obtaining the geographical position of each path to reduce or eliminate congestion, and at the same time ensure the path 1 and path 2 of the important service. The data flow remains normal.

3A is a schematic diagram of a tunnel network scenario of a network congestion processing method applied to a network using a tunneling technology according to Embodiment 3 of the present invention; FIG. 3B is a schematic flowchart of a network congestion processing method according to Embodiment 3 of the present invention. As shown in FIG. 3A, in a tunnel network, a tunnel is composed of a tunnel ingress node, a tunnel intermediate node, and a tunnel egress node. The tunnel ingress node is also the border node of the tunnel, and the data sending end, that is, the terminal UE, for example, the user host sends The data stream enters the tunnel through the node, that is, enters the network; the tunnel intermediate node refers to the network setting between the tunnel entrance node and the tunnel exit node, such as a router; the tunnel exit node is also the boundary node of the tunnel, and can receive the middle of the tunnel. a data stream forwarded by the node, and the data stream flows through the node "outflow" tunnel, that is, the outbound network, where the tunnel entry node and the egress node may be bases The eNode can be a serving gateway S-GW. Specifically, for the uplink data, the base station is a tunnel ingress node, and the serving gateway is an egress node, and for downlink data, the serving gateway is an ingress node, and the base station is an egress node. The intermediate node is a router connected between the base station and the serving gateway; in the tunnel network, there is also a core network device and a Packet Data Network (PDN) gateway P-GW. The following takes the above data as an example to describe the processing of network congestion in the tunnel network:

 Step 301: The base station eNode acquires uplink data from the terminal UE, and sends the uplink data to the serving gateway S-GW through the tunnel 1 (tunnel1);

 Step 302: The intermediate node in the tunnel 1, that is, the router detects that the tunnel 1 has network congestion, marks the data to be sent, and marks the uplink data sent to the serving gateway S-GW as congestion data;

 Step 303: When receiving the congestion data, the serving gateway S-GW collects statistics on the number of the congested data packets to obtain the congestion amount.

 Step 304: The serving gateway S-GW sends the obtained congestion amount and the tunnel 1 as congestion information to the PDN gateway P-GW;

 Step 305: The PDN gateway P-GW can adjust the transmission parameter of the tunnel 1 according to the congestion quantity to adjust data traffic entering the network from the terminal UE, and control the data flow.

 A person skilled in the art can understand that the terminal UE can establish a tunnel by the PDN gateway P-GW before the data is uploaded, that is, the tunnel 1 is established by the PDN gateway P-GW, which can determine the data stream transmission rate of the tunnel. That is, the tunnel 1 is a data transmission path composed of the base station eNode, the router, and the monthly service gateway S-GW. Therefore, the data transmission rate in the tunnel 1 can be controlled by adjusting the relevant parameters of the tunnel 1 through the PDN gateway P-GW. At the same time, the serving gateway S-GW and the PDN gateway P-GW establish a data transmission channel through the tunnel 2, and the tunnel 2 is also established by the P-GW, and is composed of the S-GW, the core network device, and the P-GW. Transmission path. The specific implementation is the same as or similar to the traditional tunnel technology, and details are not described herein again.

 In the foregoing steps 301 and 302, after the tunnel 1 is established, the uplink data uploaded by the terminal UE is transmitted in the specific tunnel 1. In this process, the router can perform network congestion detection in real time and detect the tunnel. 1 When congestion occurs, the data packet can be congested, and the same method as the above method embodiment can be used.

In step 303 and step 304 above, the serving gateway S-GW can collect statistics in the tunnel 1 in real time. The number of congested packets sent by the router, the amount of congestion obtained, and the amount of congestion and tunnels

The information is sent as congestion information to the PDN gateway P-GW, which includes the address of the serving gateway S-GW as the egress node and the router as the intermediate node.

 In the foregoing step 305, after obtaining the congestion information, the PDN gateway P-GW can control the data stream transmission rate of the tunnel 1 by modifying the relevant parameters of the tunnel 1, thereby limiting the data traffic entering the tunnel 1 through the base station, and reducing or eliminating the tunnel. 1 congestion.

 A person skilled in the art can understand that for the downlink data, that is, the data stream sent from the serving gateway S-GW to the base station, the network congestion control process is similar to the process shown in FIG. 3B, except that the base station acts as an egress node to the amount of congestion. Statistics are performed and the obtained congestion information is sent to the PDN gateway P-GW.

 It can be seen that in the tunnel application scenario shown in FIG. 3A, the PDN gateway P-GW can process the congestion information and adjust the tunnel parameters to control the data traffic entering the tunnel from the ingress node.

 FIG. 4 is a schematic flowchart diagram of a network congestion processing method according to Embodiment 4 of the present invention. The executor of the method in this embodiment may be an egress node or a congestion information concentrating node in the network. In this embodiment, the egress node may perform statistics on the congested data packets sent by the intermediate node to obtain network congestion information. As shown in FIG. 4, the method in this embodiment may include the following steps:

 Step 401: The egress node receives the congested data packet sent by the intermediate node, where the congested data packet is a data packet that marks the forwarded data packet when the intermediate node generates congestion.

 Step 402: The egress node determines the amount of congestion according to the received number of the congested data packets. Step 403: The egress node sends the network congestion information to the ingress node, where the network congestion information includes: the congestion quantity, the egress node Address and the address of the intermediate node, so that the ingress node determines the path of the network congestion based on the network congestion information, and controls the data flow that enters the network through the ingress node and passes through the path to reduce data passing through the intermediate node and the egress node. flow.

 In this embodiment, the egress node in the network can obtain the congestion amount and feed it back to the network congestion processing node or the ingress node to control the data flow entered through the ingress node to reduce or eliminate congestion.

This embodiment can be applied to a network using a tunneling technology, so that when a tunnel is congested, The data flow entering the tunnel is controlled.

 FIG. 5 is a schematic structural diagram of a network node according to Embodiment 5 of the present invention. As shown in FIG. 5, the network node in this embodiment may include a receiving module 1 1 and a processing module 12, where:

 The receiving module 11 is configured to receive network congestion information sent by the egress node of the network, where the network congestion information includes an address of the egress node, an amount of congestion, and an address of the intermediate node where congestion occurs; and the processing module 12 is configured to receive according to the receiving module 11 The address of the egress node and the address of the intermediate node in the network congestion information, determining a path of network congestion, and determining, according to the path, a data flow that needs to pass through the path in the data flow entering the network, and according to the The congestion amount, the user subscription information, and the network control policy control the data flow that needs to pass through the path to reduce data traffic passing through the intermediate node and the egress node.

 The network node provided in this implementation may be an ingress node or a network congestion processing node deployed in the network, which can process the congestion information, so that the ingress node controls the data flow that causes the network to be congested. For the specific implementation, refer to the foregoing invention. The description of the method embodiments is omitted here.

 In the embodiment shown in FIG. 5, the network node may be an ingress node in the network, and the processing module 12 may be specifically configured to determine, according to the user subscription information, a service corresponding to each data flow that needs to pass the path. Determining, according to the network control policy, the priority of the service corresponding to each data flow that needs to pass the path, and correspondingly selecting the service with a lower priority according to the congestion amount in each data flow that needs to pass the path The data flow is controlled. The foregoing processing module 12 is specifically configured to determine a priority of a service corresponding to the data flow passing the path according to a priority of a geographic location of the path set in the network control policy. In addition, the foregoing processing module 12 is specifically configured to control, according to the congestion amount, a transmission rate of all data flows that need to pass through the path, or restrict or prohibit transmission of at least a part of the data flow.

 In this embodiment, the network node may be a network node in a network using the tunneling technology. For details, refer to the description of the embodiment shown in FIG. 3A and FIG. 3B.

 FIG. 6 is a schematic structural diagram of a network node according to Embodiment 6 of the present invention. As shown in FIG. 6, the network node in this embodiment may specifically include a receiving module 21, a processing module 22, and a sending module 23, and the towel:

The receiving module 21 is configured to receive a congestion data packet sent by the intermediate node, where the congestion data packet is a packet for marking the forwarded data packet when the intermediate node generates congestion; the processing module 22, configured to determine a congestion amount according to the number of the congestion data packets received by the receiving module 21, and obtain network congestion information The network congestion information includes: the congestion amount, an address of the egress node, and an address of the intermediate node;

 The sending module 23 is configured to send, to the ingress node of the network, the network congestion information obtained by the processing module 22, so that the ingress node determines a path of network congestion based on the network congestion information and passes the ingress node. Enter the network and control the data flow through the path. The network node in this embodiment may be an egress node or a congestion information gathering node in the network, which may collect statistics on the congested data packets forwarded by the intermediate nodes in the network, and obtain congestion information. For the specific implementation, refer to the method of the present invention. The description of the example will not be repeated here.

 The network node in this embodiment may be a network node in a network using the tunnel technology. FIG. 7 is a schematic structural diagram of a network node according to Embodiment 7 of the present invention. As shown in FIG. 7, the network node of this embodiment may include a receiver 101, a memory 102, and a processor 103 connected to the memory 102. The memory 102 stores a set of program codes, and the receiver 101 and the memory 102 pass through. The bus is coupled to the processor 103, wherein:

 The receiver 101 is configured to receive network congestion information sent by an egress node of the network, where the network congestion information includes an address of the egress node, a congestion amount, and an address of an intermediate node where congestion occurs;

 The processor 103 is configured to execute the program code stored in the memory 101, to determine a path of network congestion according to an address of an egress node and an address of an intermediate node in the network congestion information received by the receiver 101, And determining, according to the path, a data flow that needs to pass through the path in the data flow entering the network, and controlling, according to the congestion amount, the user subscription information, and the network control policy, the data flow that needs to pass the path .

 In this embodiment, the processor 103 is specifically configured to determine, according to the user subscription information, a service corresponding to each data flow that needs to pass the path, and determine, according to the network control policy, a data flow that needs to pass through the path. The priority of the service, and in each data flow that needs to pass through the path, according to the congestion amount, the data flow corresponding to the service with low priority is preferentially controlled.

In this embodiment, the processor 103 is specifically configured to determine, according to a priority of a geographic location of a path set in the network control policy, a service corresponding to the data flow passing the path. Priority of service.

 In this embodiment, the processor 103 is specifically configured to control, according to the congestion amount, a transmission rate of all data flows that need to pass the path, or limit or prohibit the transmission rate of at least a part of the data flows. send.

 The network node in this embodiment can implement the foregoing steps of the method embodiment of the present invention to implement network congestion control. For the specific implementation, refer to the description of the foregoing method embodiment of the present invention.

 FIG. 8 is a schematic structural diagram of a network node according to Embodiment 8 of the present invention. As shown in FIG. 8, the network node of this embodiment may include a receiver 201, a transmitter 202, a memory 203, and a processor 204 connected to the memory 203. The memory 203 stores a set of program codes, a receiver 201, The transmitter 202 and the memory 203 are all connected to the processor 204 via a bus, wherein:

 The receiver 201 is configured to receive a congestion data packet sent by an intermediate node, where the congestion data packet is a data packet that marks the forwarded data packet when the intermediate node generates congestion;

 The processor 204 is configured to execute the program code stored in the memory 203 to determine a congestion amount according to the number of the congestion data packets received by the receiver 201, and obtain network congestion information, where the network The congestion information includes: the congestion amount, an address of the egress node, and an address of the intermediate node;

 The transmitter 202 is configured to send the network congestion information obtained by the processor 204 to an ingress node of the network, so that the ingress node determines a path of network congestion based on the network congestion information and passes the The ingress node enters the network and is controlled by the data flow of the path.

 In this embodiment, the network node may collect network congestion information and report it to the ingress node of the network, so that the ingress node implements network congestion control based on the network congestion information. For the specific implementation, refer to the description of the foregoing method embodiment of the present invention.

 FIG. 9 is a schematic structural diagram of a network system according to Embodiment 9 of the present invention. As shown in FIG. 9, the network system of this embodiment may include an ingress node 10, an egress node 20, and an intermediate node 30. The ingress node 10 may be the network node shown in FIG. 5 or FIG. 7 above, and the egress node 20 may be For the specific configuration of the network node shown in FIG. 6 or FIG. 8 , refer to the description of the foregoing network node embodiments, and details are not described herein again.

In this embodiment, the egress node 20 can obtain the congestion data packet sent by the inter-concentration node 30. The network congestion information is taken, and the network congestion information is sent to the ingress node 10, so that the ingress node 10 can obtain the path of the network congestion based on the network congestion information, and control the data flow of the path that is congested by the network.

 It can be understood by those skilled in the art that the ingress node 10 can be connected to multiple intermediate nodes 30 at the same time. Correspondingly, the egress node 20 can also be connected to multiple intermediate nodes 30, that is, in an actual network, one ingress node can simultaneously pass multiple A communication connection is established between the intermediate node and the egress node. In addition, an intermediate node can also be connected to multiple ingress nodes or multiple egress nodes at the same time.

 A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The foregoing steps include the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

 It should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art The technical solutions described in the foregoing embodiments may be modified, or some or all of the technical features may be equivalently replaced; and the modifications or substitutions do not deviate from the technical solutions of the embodiments of the present invention. range.

Claims

claims 1. A network congestion processing method, characterized by including: The entry node of the network receives the network congestion information sent by the exit node. The network congestion information includes the address of the exit node, the amount of congestion, and the address of the intermediate node where congestion occurs; Determine the path of network congestion based on the address of the exit node and the address of the intermediate node; Determine the data flows that need to pass through the path among the data flows entering the network according to the path; Control the data flow that needs to pass through the path according to the congestion amount, user subscription information and network control policy. 2. The method according to claim 1, wherein the controlling the data flow that needs to pass through the path according to the congestion amount, user subscription information and network control policy includes: Determine the service corresponding to each data flow that needs to pass through the path according to the user subscription information, and determine the priority of the service corresponding to each data flow that needs to pass through the path based on the network control policy; Among the data flows that need to pass through the path, according to the congestion amount, the data flow corresponding to the low-priority service is preferentially controlled. 3. The method according to claim 2, characterized in that: determining the priority of services corresponding to each data flow that needs to pass through the path based on the network control policy includes: according to the settings in the network control policy The priority of the geographical location of a certain path is determined, and the priority of the service corresponding to the data flow passing through the path is determined. 4. The method according to claim 1 or 2 or 3, characterized in that: controlling the data flow that needs to pass through the path according to the congestion amount, user subscription information and network control policy includes: According to the congestion amount, the sending rate of all data flows that need to pass through the path is controlled, or the sending rate of at least a part of the data flows is limited or prohibited. 5. A network congestion processing method, characterized by including: The egress node of the network receives the congestion data packet sent by the intermediate node, where the congestion data packet is the data packet used to mark the forwarded data packet when congestion occurs at the intermediate node; Determine the amount of congestion based on the number of received congestion data packets; Send network congestion information to an entry node of the network, where the network congestion information includes: the congestion amount, the address of the exit node, and the address of the intermediate node, so that the entry node determines based on the network congestion information The network is congested on a path and controls the flow of data entering the network through the entry node and passing through the path. 6. A network node, characterized by including: A receiving module, configured to receive network congestion information sent by an exit node of the network, where the network congestion information includes the address of the exit node, the amount of congestion, and the address of the intermediate node where congestion occurs; A processing module, configured to determine the path of network congestion based on the address of the exit node and the address of the intermediate node in the network congestion information received by the receiving module, and determine based on the path that the data flow entering the network needs to pass through The data flow of the path is controlled, and the data flow that needs to pass through the path is controlled according to the congestion amount, user subscription information and network control policy. 7. The network node according to claim 6, characterized in that, the processing module is specifically configured to determine the service corresponding to each data flow that needs to pass through the path according to the user subscription information, and control based on the network The policy determines the priority of the services corresponding to each data flow that needs to pass through the path, and among each data flow that needs to pass through the path, based on the congestion amount, the data flow corresponding to the low-priority service is given priority to be controlled. . 8. The network node according to claim 7, characterized in that the processing module is specifically configured to determine the data flow passing through the path according to the priority of the geographical location of the path set in the network control policy. The priority of the corresponding business. 9. The network node according to claim 6, 7 or 8, characterized in that the processing module is specifically configured to control the sending rate of all data flows that need to pass through the path according to the congestion amount, Or limit the sending rate of at least part of the data stream or prohibit sending. 10. A network node, characterized in that it includes a receiver, a memory and a processor connected to the memory, and a set of program codes is stored in the memory, wherein: The receiver is configured to receive network congestion information sent by an exit node of the network. The network congestion information includes the address of the exit node, the amount of congestion, and the location of the intermediate node where congestion occurs. address; The processor is configured to execute the program code stored in the memory to determine the path of network congestion based on the address of the exit node and the address of the intermediate node in the network congestion information received by the receiver, and based on the The path determines the data flow that needs to pass through the path among the data flows entering the network, and controls the data flow that needs to pass through the path according to the congestion amount, user subscription information and network control policy. 11. The network node according to claim 10, characterized in that, the processor is specifically configured to determine the service corresponding to each data flow that needs to pass through the path according to the user subscription information, and control based on the network The policy determines the priority of the services corresponding to each data flow that needs to pass through the path, and among each data flow that needs to pass through the path, based on the congestion amount, the data flow corresponding to the low-priority service is given priority to be controlled. . 12. The network node according to claim 11, wherein the processor is specifically configured to determine the data flow passing through the path according to the priority of the geographical location of the path set in the network control policy. The priority of the corresponding business. 13. The network node according to claim 10, 11 or 12, characterized in that the processor is specifically configured to control the sending rate of all data flows that need to pass through the path according to the congestion amount, Or limit the sending rate of at least part of the data stream or prohibit sending. 14. A network node, characterized by: including: A receiving module, configured to receive congestion data packets sent by the intermediate node, where the congestion data packets are data packets used to mark forwarded data packets when congestion occurs at the intermediate node; A processing module, configured to determine the congestion amount according to the number of the congestion data packets received by the receiving module, and obtain network congestion information. The network congestion information includes: the congestion amount, the address of the exit node, and the The address of the intermediate node; A sending module, configured to send the network congestion information obtained by the processing module to an entry node of the network, so that the entry node determines the path of network congestion based on the network congestion information and corrects the traffic entering the network through the entry node. And the data flow passing through the path is controlled. 15. A network node, characterized in that it includes a receiver, a transmitter, a memory, and a processor connected to the memory. A set of program codes is stored in the memory, wherein: the receiver is used to receive intermediate The congestion data packet sent by the node, the congestion data packet A data packet that marks the forwarded data packet when congestion occurs at the intermediate node; the processor is configured to execute the program code stored in the memory to respond to the congestion data packet received by the receiver. number, determine the amount of congestion, and obtain network congestion information. The network congestion information includes: the amount of congestion, the address of the exit node, and the address of the intermediate node; The transmitter is configured to send the network congestion information obtained by the processor to an entry node of the network, so that the entry node determines a path of network congestion based on the network congestion information and corrects the path passing through the entry node. The flow of data entering the network and passing through the path is controlled. 16. A network system, characterized in that it includes: an entry node, an intermediate node and an exit node, wherein the entry node is a network node according to any one of claims 6-13, and the exit node is a Using the network node according to claim 14 or 15.