WO2017080301A1

WO2017080301A1 - Method for transferring network exit congestion state, network exit selection method, and router

Info

Publication number: WO2017080301A1
Application number: PCT/CN2016/098823
Authority: WO
Inventors: 李振强
Original assignee: 中国移动通信集团公司
Priority date: 2015-11-13
Filing date: 2016-09-13
Publication date: 2017-05-18
Also published as: CN106713161A

Abstract

Provided are a method for transferring a network exit congestion state in real time, a network exit selection method, and a router. The method for transferring a network exit congestion state in real time comprises: detecting a network exit congestion state in real time; and when the network exit congestion state reaches a pre-determined condition, declaring a route to a border gateway protocol (BGP) neighbour, the declared route being marked with a network exit congestion state having reached the pre-determined condition. In the present solution, a network exit congestion state is detected in real time, and when the network exit congestion state reaches a pre-determined condition, a route is declared to a border gateway protocol (BGP) neighbour, the declared route being marked with a network exit congestion state having reached the pre-determined condition, so that a router (in a provincial network) which needs to use a network exit state can sense the network exit congestion situation in real time, then traffic can be dynamically dredged between a plurality of network exits according to a local policy based on the received network exit congestion state, thus ensuring the user experience, and improving the utilization efficiency of the network exits.

Description

Method for transmitting network egress congestion state, network egress selection method, and router

Cross-reference to related applications

The present application claims priority to Chinese Patent Application No. 201510778556.8, filed on Jan. 13, 2015, in

Technical field

The present disclosure relates to the field of communication technologies, and in particular, to a method for transmitting a network egress congestion state in real time, a network egress selection method, and a router.

Background technique

As shown in Figure 1, there are often multiple egresses in the network that can reach content sources outside the carrier's network. If the egress is congested when users on the carrier network access the e-network resources, the user's access experience will be seriously affected.

In the related technology, the provincial network cannot be perceived in real time due to the congestion of the backbone network of the operator. In order to improve the user experience, the operator's provincial network will channel the traffic of some important customers through third-party exports through policy routing. However, the congestion of the third-party export, the operator's provincial network is not obtained in real time, usually by purchasing enough bandwidth to ensure that after purchasing a certain bandwidth, the operator's provincial network must ensure that the traffic dispatched to the third-party export does not exceed the purchase bandwidth. . If the third-party export congestion occurs due to traffic bursts, etc., the operator's provincial network cannot be perceived in real time, and can only be discovered when a complaint is received or the network quality is checked.

That is, the operator's provincial network cannot sense the congestion of the network exit in real time, and the traffic cannot be scheduled in real time according to the congestion situation among multiple outlets, which will affect the user's Internet access experience and affect the bandwidth of the operator's network. Use efficiency.

Summary of the invention

The purpose of the present disclosure is to provide a method for real-time delivery of a network egress congestion state, a network egress selection method, and a router, which solves the problem that the provincial network cannot perceive the network egress congestion condition in real time.

In order to solve the above technical problem, the embodiment of the present disclosure provides a method for real-time delivery of a network egress congestion state, which is applied to a first router, including:

Real-time detection of network exit congestion status;

When the network egress congestion state reaches a predetermined condition, the route is announced to the border gateway protocol BGP neighbor;

Wherein, the declared route has a network exit congestion state that reaches the predetermined condition.

Optionally, before the routing the BGP neighbor to the border gateway protocol, the method further includes:

When the network egress congestion state reaches the predetermined condition, the route is generated according to all links or partial links that trigger route advertisement.

Optionally, the step of generating the route according to all links or partial links that trigger the route advertisement includes:

The network egress congestion state that meets the predetermined condition is identified by the community attribute community, where the network egress congestion state includes a bandwidth of the network egress link and a current link utilization rate.

The network egress congestion state that meets the predetermined condition is identified by the extended community attribute extended community, where the network egress congestion state includes a bandwidth of the network egress link and a current link utilization rate.

Optionally, the network egress congestion state further includes a network egress link delay and a link stability.

Optionally, the step of declaring a route to the border gateway protocol BGP neighbor when the network egress congestion state reaches a predetermined condition includes:

When the link utilization of the network egress changes, or the link utilization changes exceed the first threshold, or the link utilization reaches the second threshold, the route is announced to the border gateway protocol BGP neighbor.

Optionally, the step of advertising the route to the border gateway protocol BGP neighbor includes:

Advertise routes to all BGP neighbors or some BGP neighbors.

The present disclosure also provides a first router, including:

a detecting module, configured to detect a network export congestion state in real time;

An announcement module, configured to announce a route to a border gateway protocol BGP neighbor when the network egress congestion state reaches a predetermined condition;

The present disclosure also provides a network egress selection method, which is applied to a second router, including:

Receiving a route advertised by the first router; wherein the first router and the second router are border gateway protocol BGP neighbors, and the route is information generated by the first router according to a network egress congestion state that reaches a predetermined condition ;

The network exit selection is performed by using the network egress congestion state in the route in combination with the local policy.

Optionally, the step of using the network egress congestion state in the route to perform network egress selection in combination with the local policy includes:

When the link utilization of the backbone network egress is lower than the third threshold, the egress network does not exceed the outbound traffic of the remaining available bandwidth of the backbone network egress.

Optionally, before the utilizing the network egress congestion state in the route and the local policy to perform the network egress selection, the network egress selection method further includes:

The configuration generates the local policy.

Optionally, the local policy includes: selecting a network exit with higher stability when the remaining available bandwidth of the network egress is the same.

The present disclosure also provides a second router, including:

a receiving module, configured to receive a route advertised by the first router, where the first router and the second router are border gateway protocol BGP neighbors, and the route is a network exit by the first router according to a predetermined condition Information generated by the congestion state;

The processing module is configured to perform network egress selection by using a network egress congestion state in the route in combination with a local policy.

The beneficial effects of the above technical solutions of the present disclosure are as follows:

In the foregoing solution, the method for real-time delivery network egress congestion state detects the congestion state of the network egress in real time; when the network egress congestion state reaches a predetermined condition, announces a route to the border gateway protocol BGP neighbor; and the announced route identifier is reached. The network egress congestion state of the predetermined condition enables the second router - the router (provincial network) that needs to use the network egress state to be able to sense the network egress congestion status in real time, and then according to the received network egress congestion status according to the local policy in multiple Dynamically diverting traffic between network egresses ensures user experience and improves network export utilization efficiency.

DRAWINGS

1 is a schematic diagram of a network architecture in the related art;

2 is a schematic flowchart of a method for real-time delivery of a network egress congestion state according to some embodiments of the present disclosure;

3 is a schematic flowchart of a method for selecting a network egress according to some embodiments of the present disclosure;

4 is a schematic diagram of a network architecture according to some embodiments of the present disclosure;

FIG. 5 is a schematic diagram of an extended community definition according to some embodiments of the present disclosure; FIG.

6 is a schematic diagram of an example of an extended community encoding according to some embodiments of the present disclosure;

FIG. 7 is a schematic structural diagram of a first router according to some embodiments of the present disclosure;

FIG. 8 is a schematic structural diagram of a second router according to some embodiments of the present disclosure;

FIG. 9 is a schematic diagram of an example of an extended community encoding according to some embodiments of the present disclosure.

detailed description

The technical problems, the technical solutions, and the advantages of the present invention will be more clearly described in conjunction with the accompanying drawings and specific embodiments.

The present disclosure provides various solutions for the problem that the provincial network cannot perceive the network export congestion status in real time in the related art, and the specific solutions are as follows:

As shown in FIG. 2, a method for real-time delivery network egress congestion status provided by some embodiments of the present disclosure is applicable to a first router (egress router), where the method includes:

Step 21: Real-time detection of network exit congestion status;

Step 22: Declare a route to the border gateway protocol BGP neighbor when the network egress congestion state reaches a predetermined condition;

Some embodiments of the present disclosure detect the congestion state of the network egress in real time; when the network egress congestion state reaches a predetermined condition, announce a route to the border gateway protocol BGP neighbor; and the announced route flag has a network egress congestion state that reaches a predetermined condition, so that A router (provincial network) that needs to use the network egress status can sense the network egress congestion status in real time, and thus can be based on the received network. The export congestion status dynamically diverts traffic between multiple network outlets according to local policies, ensuring user experience and improving network export utilization efficiency.

Further, before the advertised route to the border gateway protocol BGP neighbor, the method further includes: determining, in real time, whether the congestion status reaches the predetermined condition.

Further, before the routing of the BGP neighbor to the border gateway protocol, the method further includes: when the network egress congestion state reaches the predetermined condition, generating according to all links or partial links that are triggered by the triggering route The route.

As with the manner in which the network exit congestion state that meets the predetermined condition is identified, some embodiments of the present disclosure provide two measures:

In a first measure, the step of generating the route according to all links or partial links that are triggered by the triggering route includes: identifying, by using a community attribute community, a network egress congestion state that meets the predetermined condition, where the network egress congestion state includes The bandwidth of the network egress link and the current utilization of the link.

The second measure, the step of generating the route according to all links or partial links that trigger the route advertisement includes: identifying, by using an extended community attribute extended community, a network egress congestion state that reaches the predetermined condition, where the network egress is congested The status includes the bandwidth of the network egress link and the current utilization of the link.

When the extended community is used to deliver the network egress congestion state, because the field space is large enough, the egress link delay and stability can be carried while the network egress bandwidth and the link current utilization are transmitted. Information such as the number of times the link is interrupted, etc.; that is, the network egress congestion state may also include network egress link delay and link stability.

Specifically, when the network egress congestion state reaches a predetermined condition, the step of declaring a route to the border gateway protocol BGP neighbor includes: changing a link utilization rate at the network egress, or changing the link utilization rate by more than the first When the threshold, or link utilization, reaches the second threshold, the route is announced to the Border Gateway Protocol BGP neighbor.

Optionally, the step of advertising the route to the border gateway protocol BGP neighbor comprises: declaring a route to all BGP neighbors or part of the BGP neighbors. A BGP neighbor can be either a router or a route reflector RR.

As shown in FIG. 3, a network egress selection method provided by some embodiments of the present disclosure may be applied to a second router (a router that needs to use a network egress state), where the network egress selection method includes:

Step 31: Receive a route advertised by the first router, where the first router and the second router are border gateway protocol BGP neighbors, and the route is a network egress congestion state by the first router according to a predetermined condition. Generated information;

Step 32: Perform network egress selection by using the network egress congestion state in the route in combination with the local policy (and further divert different traffic through different network outlets).

The step of receiving the route advertised by the first router may be directly receiving the route advertised by the first router, or the route advertised by the first router forwarded by the route reflector RR.

Some embodiments of the present disclosure ensure the user experience and improve the network by receiving the route advertised by the first router and utilizing the network egress congestion state in the route in combination with the local policy for network egress selection, and then diverting different traffic through different network outlets. Export utilization efficiency.

The preset condition includes that the link utilization of the network egress changes, or the link utilization change exceeds a first threshold, or the link utilization reaches a second threshold.

Optionally, the step of utilizing the network egress congestion state in the route and the local policy to perform network egress selection includes: the link utilization at the backbone network egress is lower than a third threshold (such as 80%), or the remaining bandwidth When the value is greater than the fourth threshold (for example, 2G), the egress network does not exceed the remaining available bandwidth of the backbone network outlet.

Further, before the utilizing the network egress congestion state in the route and the local policy to perform the network egress selection, the network egress selection method further includes: configuring to generate the local policy.

Specifically, the local policy includes: selecting a network egress with higher stability or less delay when the remaining available bandwidth of the network egress is the same.

Some embodiments of the present disclosure, in conjunction with the foregoing embodiments, describe a method for transmitting and receiving congestion status of a real-time delivery network from a transmitting and receiving side, and a network egress selection method.

As shown in FIG. 4, some embodiments of the present disclosure provide a network egress router and a router that needs to use a network egress state based on an existing network system, and a network egress router and a router that needs to use a network egress state, and Between BGP routers, two routers that need to use the network egress state communicate through iBGP (Internal Border Gateway Protocol); In addition, both parties communicate via eBGP (External Border Gateway Protocol).

The network egress router in the carrier backbone network is used to sense the egress status of the backbone network, and the BGP router in the carrier backbone network is used to forward the route announced by the network egress router in the carrier backbone network;

After receiving the route forwarded by the BGP router, the router in the carrier's provincial network needs to use the network state router to forward the route to the router in the operator's provincial network and the router that needs to use the network state. The router learns according to the The route, and the network egress status of the third-party egress that it perceives, the traffic is diverted, and the router announces the network egress status of the third-party egress that it perceives to the router in the operator's provincial network that needs to use the network state. The solution flow provided by some embodiments of the present disclosure is specifically described below.

sender

First, the egress router needs to detect the congestion status of the network egress, and announce the route to the BGP neighbor according to the degree of change of the congestion status. The announced route identifies the latest congestion status of the egress. The corresponding egress router needs to have the following functional modules: a network egress congestion state detection module, a network egress congestion state change degree decision module, a network exit congestion state indication generation module, and a route advertisement module.

Specifically, the route is advertised to the BGP neighbor according to the degree of change of the congestion state, which may include the following three situations:

1. The routing announcement is triggered as long as the link utilization of the egress changes.

2. The route advertisement is triggered only when the change of the egress link utilization exceeds a certain threshold (for example, 10%, and the specific threshold can be set);

3. The route advertisement is triggered only when the egress link utilization reaches a set threshold (for example, more than 80% or less than 50%, and the specific threshold can be set).

Optionally, the route is advertised to the BGP neighbor, which may include the following four situations:

1. Can advertise routes to all BGP neighbors;

2. You can also advertise routes to only a part of BGP neighbors through configuration.

3. The announced route may be all routes learned from the link that triggered the route advertisement;

4. It is also possible to configure only a part of the route learned from the link that triggers the route advertisement (such as the route of some or some community, the route learned from a certain operator, etc.).

In addition, the manner in which the announced route indicates the latest congestion status of the exit may include the following two types:

1. Identify the latest congestion status of the exit through the community (RFC1997): the community is a four-byte unsigned integer, usually written in the form of an AS number: a 16-bit integer. In some embodiments of the present disclosure, a reserved or private AS number may be used to indicate that the community is used to indicate the egress link congestion status, and the last 16 digits indicate the specific congestion status of the egress link, which is a 16-digit number. Specific coding methods include but are not limited to:

The upper 8 bits indicate the bandwidth of the egress link in Gbps; the lower 8 bits indicate the current utilization of the link in %.

2. Identify the latest congestion state of the exit through the extended community (RFC4360): as shown in Figure 5, the extended community is a octet with a certain structure, and the highest one or two bytes indicates the type of the extended community. The other bytes are specific values. Some embodiments of the present disclosure may define a new type (the specific type value is determined by standardization work) to indicate that the extended community carries the egress link congestion status information. The specific coding method of the extended community value part includes but is not limited to:

The last byte of the value part indicates the current utilization of the egress link in %; the other bytes of the value part represent the link bandwidth of the egress, in Gbps. A possible coding example is shown in Figure 9. A subtype can be applied from the Transitive Four-Octet AS-Specific Extended Community (RFC7153) to represent the extended community. Attributes are used to convey the state of network egress congestion. 0x02 in FIG. 9 indicates that this is a transferable 4-byte AS-specific extended community attribute; the sub-type in FIG. 9 indicates that the extended community attribute is used to convey the network exit congestion state; Sender AS Number in FIG. (Sender Autonomous Area Number) is the autonomous domain number of the device that generates the extended community attribute, and may be a 4-byte autonomous domain number or a 2-byte autonomous domain number. When the autonomous domain number of the device that generates the extended community attribute is 2 bytes, according to the requirements of RFC6793, the highest two bytes of the Sender AS Number field may be set to 0, and the lower two bytes are stored in the device. 2 byte autonomous domain number. The seventh byte is the egress bandwidth, in Gbps, the actual physical bandwidth of the network egress link is recorded; the eighth byte is the egress utilization, in %, and records the current utilization information of the egress link of the network.

In addition, when the extended community is used to deliver the network egress congestion state, because the field space is large enough, the egress outlet bandwidth and the current utilization of the link can be carried while carrying the egress. Link delay, stability (link stability can be characterized by the number of link interruptions in a day, etc.), etc., a possible coding example is shown in Figure 6.

The first two bytes are the type of the extended community, and the specific value is determined by standardization, indicating that the extended community carries the network egress congestion state; the third and fourth bytes are the exit stability, and the record link is in the middle of the day. The number of interruptions; the fifth byte is the exit delay, in ms, recording the current delay information of the network egress link; the sixth and seventh bytes are the egress bandwidth, in Gbps, recording the actual physical bandwidth of the network egress link; The 8th byte is the export utilization rate, in %, and records the current utilization information of the network egress link.

receiver

After receiving the BGP route advertisement that carries the egress congestion status, the router that needs to use the egress status of the egress withdraws relevant information (carrying network egress congestion status information) from the community or the extended community, and uses the information according to the local policy to perform export selection. When the remaining available bandwidth of the outlet is the same, an outlet with high export stability or small delay is selected, and different flows are diverted through different outlets.

Correspondingly, the router needs to have the following functional modules: a BGP route receiving module, a network egress congestion state community or an extended community processing module, and a traffic grooming policy implementation module.

Specifically, comprehensively utilizing the information according to the local policy for export selection may include:

When the link utilization rate of the backbone network exit is lower than a certain threshold (for example, 80%), or the remaining bandwidth is greater than a certain threshold (for example, 2G), the outbound traffic that does not exceed the remaining available bandwidth of the backbone export is preferentially routed from the backbone exit.

As can be seen from the above, some embodiments of the present disclosure provide a solution that enables the provincial network to dynamically divert traffic between multiple egresses according to the received egress congestion status according to the local policy, thereby ensuring user experience and improving network egress utilization efficiency.

As shown in FIG. 7, some embodiments of the present disclosure provide a first router, which can be used to transmit a network egress congestion status in real time, where the first router includes:

The detecting module 71 is configured to detect a network outlet congestion state in real time;

The declaring module 72 is configured to announce a route to the border gateway protocol BGP neighbor when the network egress congestion state reaches a predetermined condition;

Some embodiments of the present disclosure detect the congestion state of the network egress in real time; when the network egress congestion state reaches a predetermined condition, announce a route to the border gateway protocol BGP neighbor; and the announced route flag has a network egress congestion state that reaches a predetermined condition, so that A router (provincial network) that needs to use the network egress state can sense the network egress congestion status in real time, and can dynamically divert traffic between multiple network egresses according to the received network egress congestion status according to the local policy, thereby ensuring the user experience and improving the user experience. Network export utilization efficiency.

Further, the first router further includes: a determining module, configured to determine, in real time, whether the congestion state reaches the predetermined condition before the announcement module announces the route to the border gateway protocol BGP neighbor.

Further, the first router further includes: a generating module, configured to: before the announcement module announces the route to the border gateway protocol BGP neighbor, when the network egress congestion state reaches the predetermined condition, according to the trigger routing announcement All links or partial links generate the route.

In a first measure, the generating module includes: a first identifier submodule, configured to identify, by using a community attribute community, a network egress congestion state that meets the predetermined condition, where the network egress congestion state includes a bandwidth and a chain of a network egress link. Current utilization of the road.

In a second aspect, the generating module includes: a second identifier submodule, configured to identify, by using an extended community attribute extended community, a network egress congestion state that meets the predetermined condition, where the network egress congestion state includes a bandwidth of a network egress link. And the current utilization of the link.

When the extended community is used to deliver the network egress congestion state, because the field space is large enough, the egress link delay and stability can be carried while transmitting the network egress utilization (the stability of the link can be interrupted by the link in one day. Information such as the number of times, etc.; that is, the network egress congestion state may also include network egress link delay and link stability.

Specifically, the announcement module includes: a first announcement sub-module, configured to change a link utilization rate of the network egress, or when the link utilization change exceeds a first threshold, or when the link utilization reaches a second threshold Declare a route to the Border Gateway Protocol BGP neighbor.

Optionally, the announcement module includes: a second announcement sub-module, configured to all BGP neighbors or Some BGP neighbors announce routes. A BGP neighbor can be either a router or a route reflector RR.

The foregoing embodiments of the foregoing method for real-time delivery of a network egress congestion state are applicable to the embodiment of the first router, and can achieve the same technical effect.

As shown in FIG. 8, some embodiments of the present disclosure provide a second router, which can be used for network egress selection, and the second router includes:

The receiving module 81 is configured to receive a route advertised by the first router, where the first router and the second router are border gateway protocol BGP neighbors, and the route is a network that is determined by the first router according to a predetermined condition. Information generated by the export congestion status;

The processing module 82 is configured to perform network egress selection by using a network egress congestion state in the route in combination with a local policy (and further divert different traffic through different network outlets).

The receiving the route advertised by the first router may be directly receiving the route advertised by the first router, or may be the route advertised by the first router forwarded by the route reflector RR.

Optionally, the processing module includes: a processing submodule, configured to: when the link utilization of the backbone network egress is lower than a third threshold (such as 80%), or when the remaining bandwidth is greater than a fourth threshold (such as 2G), The backbone network exit grooming does not exceed the outbound traffic of the remaining available bandwidth of the backbone network exit.

Further, the second router further includes: a configuration module, configured to generate the local policy before the processing module uses the network egress congestion state in the route and the local policy to perform network egress selection.

The implementation examples of the foregoing network egress selection method are applicable to the embodiment of the second router, and the same technical effects can be achieved.

It should be noted that many of the functional components described in this specification are referred to as modules/submodules. In order to more specifically emphasize the independence of its implementation.

In embodiments of the present disclosure, modules/sub-modules may be implemented in software for execution by various types of processors. For example, an identified executable code module can comprise one or more physical or logical blocks of computer instructions, which can be constructed, for example, as an object, procedure, or function. Nonetheless, the executable code of the identified modules need not be physically located together, but may include different instructions stored in different bits that, when logically combined, constitute a module and implement the provisions of the module. purpose.

For example, some embodiments of the present disclosure provide a first router that can be used to deliver a network egress congestion status in real time, the first router including a processor, the processor is configured to:

Real-time detection of network exit congestion status;

Some embodiments of the present disclosure provide a second router, which can be used for network egress selection, and the second router includes a processor, where the processor is used to:

The network egress congestion state in the route is combined with the local policy for network egress selection (and then different traffic is diverted through different network outlets).

In practice, the executable code module can be a single instruction or a plurality of instructions, and can even be distributed across multiple different code segments, distributed among different programs, and distributed across multiple memory devices. As such, operational data may be identified within the modules and may be implemented in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set, or may be distributed at different locations (including on different storage devices), and may at least partially exist as an electronic signal on a system or network.

When the module can be implemented by software, considering the level of the existing hardware process, the module can be implemented in software, and the technician can construct a corresponding hardware circuit to implement the corresponding function without considering the cost. The hardware circuit includes conventional ultra-large scale integration (VLSI) circuits or gate arrays and existing semiconductors such as logic chips, transistors, or other discrete components. The modules can also be implemented with programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices, and the like.

The above is a preferred embodiment of the present disclosure, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present disclosure. These improvements and retouchings should also be considered. It is the scope of protection of this disclosure.

Claims

A method for real-time delivery of a network egress congestion state is applied to a first router, where the method includes:

Real-time detection of network exit congestion status;

When the network egress congestion state reaches a predetermined condition, the route is announced to the border gateway protocol BGP neighbor;

Wherein, the declared route has a network exit congestion state that reaches the predetermined condition.
The method of claim 1, wherein before the routing of the BGP neighbor to the border gateway protocol, the method further comprises:

When the network egress congestion state reaches the predetermined condition, the route is generated according to all links or partial links that trigger route advertisement.
The method of claim 2, wherein the step of generating the route according to all links or partial links that trigger a route announcement comprises:

The network egress congestion state that meets the predetermined condition is identified by the community attribute community, where the network egress congestion state includes a bandwidth of the network egress link and a current link utilization rate.
The method of claim 2, wherein the step of generating the route according to all links or partial links that trigger a route announcement comprises:

The network egress congestion state that meets the predetermined condition is identified by the extended community attribute extended community, where the network egress congestion state includes a bandwidth of the network egress link and a current link utilization rate.
The method of claim 4 wherein said network egress congestion status further comprises network egress link delay and link stability.
The method of claim 1, wherein the step of declaring a route to a Border Gateway Protocol BGP neighbor when the network egress congestion state reaches a predetermined condition comprises:

When the link utilization of the network egress changes, or the link utilization changes exceed the first threshold, or the link utilization reaches the second threshold, the route is announced to the border gateway protocol BGP neighbor.
The method of claim 1 wherein said step of declaring a route to a Border Gateway Protocol BGP neighbor comprises:

Advertise routes to all BGP neighbors or some BGP neighbors.
A first router comprising:

a detecting module, configured to detect a network export congestion state in real time;

An announcement module, configured to announce a route to a border gateway protocol BGP neighbor when the network egress congestion state reaches a predetermined condition;

Wherein, the declared route has a network exit congestion state that reaches the predetermined condition.
A network exit selection method applied to a second router, including:

Receiving a route advertised by the first router; wherein the first router and the second router are border gateway protocol BGP neighbors, and the route is information generated by the first router according to a network egress congestion state that reaches a predetermined condition ;

The network exit selection is performed by using the network egress congestion state in the route in combination with the local policy.
The network egress selection method according to claim 9, wherein the step of utilizing a network egress congestion state in the route in combination with a local policy for network egress selection comprises:

When the link utilization of the backbone network egress is lower than the third threshold, the egress network does not exceed the outbound traffic of the remaining available bandwidth of the backbone network egress.
The network egress selection method according to claim 9, wherein the network egress selection method further comprises: before the utilizing the network egress congestion state in the route and the local policy to perform network egress selection;

The configuration generates the local policy.
The network egress selection method according to claim 11, wherein the local policy comprises at least selecting a network exit with higher stability when the remaining available bandwidth of the network egress is the same.
A second router comprising:

a receiving module, configured to receive a route advertised by the first router, where the first router and the second router are border gateway protocol BGP neighbors, and the route is a network exit by the first router according to a predetermined condition Information generated by the congestion state;

The processing module is configured to perform network egress selection by using a network egress congestion state in the route in combination with a local policy.