CN114338414A

CN114338414A - Backbone network topology discovery method and device and control equipment

Info

Publication number: CN114338414A
Application number: CN202210114760.XA
Authority: CN
Inventors: 彭寒秋; 黄晓伟; 沈之光
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2022-01-30
Filing date: 2022-01-30
Publication date: 2022-04-12
Anticipated expiration: 2042-01-30
Also published as: CN114338414B

Abstract

The application provides a method, a device and a control device for discovering a backbone network topology, wherein the method comprises the following steps: acquiring detection data between areas capable of reflecting the adjacency relation of network equipment on a backbone network, wherein the detection data at least comprises the detection data between the areas meeting preset conditions; and carrying out network topology discovery according to the detection data to acquire the network topology of the backbone network. By acquiring the detection data reflecting the adjacency relation of the network equipment on the backbone network, the incidence relation between part of routing addresses on the backbone network and the physical links can be determined, and a network topology structure with the incidence relation between the routing addresses and the physical links is constructed.

Description

Backbone network topology discovery method and device and control equipment

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method, an apparatus, and a control device for discovering a topology of a backbone network.

Background

The cross-provincial data transmission of the cloud service provider depends on operator networks and private lines, but the layout of the cross-provincial data transmission is usually data which is not disclosed, the operator comprises thousands of transit nodes in a nationwide backbone network, and the combination and the dependency relationship of the nodes are complex.

The existing topology discovery method can only obtain the ip address of the router on one path, and can not determine the incidence relation between the ip address of the router and a key physical link in the backbone network topology, such as a span-provincial special line in a backbone network. Thus, when an application such as a failure needs to be performed according to the backbone network topology, it is difficult to perform accurate positioning.

Disclosure of Invention

The problem to be solved by the application is that the association relation between the router ip address and part of the physical link cannot be established in the existing backbone network topology discovery method.

In order to solve the above problem, the present application first provides a method for discovering a backbone network topology, including:

acquiring detection data between areas capable of reflecting the adjacency relation of network equipment on a backbone network, wherein the detection data at least comprises the detection data between the areas meeting preset conditions;

and carrying out network topology discovery according to the detection data to acquire the network topology of the backbone network.

In one embodiment, the inter-region satisfying the predetermined condition is a trans-provincial inter-region.

In an embodiment, the performing network topology discovery according to the probe data to obtain the network topology of the backbone network includes:

determining the foreign trunk information and the inbound trunk information based on the one-way detection data among the regions;

and constructing the network topology of the backbone network according to the one-way detection data among the areas and the corresponding outbound trunk information and inbound trunk information.

In one embodiment, the probe data includes routing address information and round trip delay for each hop of a plurality of routing links;

the determining of the foreign trunk information and the inbound trunk information based on the one-way detection data between each region includes:

splitting unidirectional detection data between each region based on hop counts, and determining a routing address information set and round-trip delay of each hop count;

and determining the hop count and the routing address data of the outbound trunk and the hop count and the routing address data of the inbound trunk according to the routing address information set and the round-trip delay of each hop count.

In an embodiment, after the splitting the unidirectional probe data between each region based on the hop count and determining the routing address information set and the round trip delay of each hop count, the method further includes:

smoothing is performed based on the round trip delay per hop count.

In one embodiment, the unidirectional detection data between the regions is detection data from a source region to a destination region;

after determining the foreign trunk information and the inbound trunk information based on the one-way detection data between each region, the method further comprises the following steps:

and correcting the foreign trunk information and the domestic trunk information between the regions based on the aggregation characteristics of the same source region and the same target region.

In one embodiment, the modifying the information of the foreign stems and the information of the home stems between the domains based on the aggregation characteristics of the same source domain and the same destination domain includes:

based on the inbound trunk information of the same source region, revising the inbound trunk information among the regions corresponding to the source region;

and correcting the foreign trunk information between the areas corresponding to the target area based on the foreign trunk information of the same target area.

In one embodiment, the modifying the inbound trunk information between the regions corresponding to the source region based on the inbound trunk information of the same source region includes:

counting the inbound trunk information of the same source region, and sequencing according to the occurrence frequency of the hop count of the inbound trunk;

determining a union set of the sequenced inbound trunk information based on the routing address information set of the inbound trunk information corresponding to each sequence;

and according to the hop count and the routing address information set between the non-head-ranked areas, correcting the trunk information between the areas based on the union and the hop count ranked in advance.

In an embodiment, the constructing a network topology of the backbone network according to the one-way probe data between the regions and the corresponding outbound trunk information and inbound trunk information includes:

according to the outbound trunk information and the inbound trunk information among the regions, dividing the one-way detection data among the regions into a source region provincial network, a target region provincial network and a national trunk network;

constructing a network topology of a backbone network according to the divided unidirectional detection data among the areas;

and based on the routing address characteristics of the physical link, determining routing address information of two ends of the physical link and routing address information of the national trunk switch unit according to the unidirectional detection data and the reverse detection data between the areas, and correcting the network topology of the backbone network.

In an embodiment, the constructing a network topology of the backbone network according to the one-way probe data between the regions and the corresponding outbound trunk information and inbound trunk information further includes:

determining a target area passing through the same inbound trunk exchange unit based on the inbound trunk information of the same source area;

determining a source region passing through the same outbound trunk exchange unit based on the outbound trunk information of the same target region;

and modifying the network topology of the backbone network.

In one embodiment, the acquiring probe data between areas on the backbone network includes:

sending a detection data acquisition request to part or all of detection nodes deployed in an area, and receiving the detection data returned by the detection nodes according to the detection data acquisition request; alternatively, the first and second electrodes may be,

and receiving the detection data sent by part or all of the detection nodes deployed in the area at a preset time point.

In an embodiment, the probe data of the probe node sends a probe packet to some or all probe nodes deployed in other areas through the probe node, and receives data returned by a link transit route based on the probe packet to obtain the data.

In one embodiment, in the acquisition of the probe data of the probe node, the probe node transmits a probe packet to some or all of the probe nodes deployed in other areas based on at least one of a TCP protocol and a UDP protocol.

In one embodiment, the area is a city.

This application is next a controlgear, it includes: a memory, a processor, and a communications component;

the memory for storing a program;

the communication component is used for acquiring detection data between areas capable of reflecting the adjacency relation of the network equipment on the backbone network, and the detection data at least comprises detection data between the areas meeting preset conditions;

the processor, coupled to the memory and the communication component, to execute the program to:

In one embodiment, the processor is specifically configured to:

the processor is specifically configured to:

the processor is further specifically configured to:

In one embodiment, the processor is specifically configured to:

In one embodiment, the communication component is specifically configured to:

The present application provides a backbone network topology discovery apparatus, which includes:

the data acquisition module is used for acquiring detection data between areas capable of reflecting the adjacency relation of the network equipment on the backbone network, wherein the detection data at least comprises detection data between the areas meeting preset conditions;

and the topology discovery module is used for carrying out network topology discovery according to the detection data so as to acquire the network topology of the backbone network.

By acquiring the detection data reflecting the adjacency relation of the network equipment on the backbone network, the method and the system can determine the incidence relation between part of routing addresses on the backbone network and the physical links, and construct a network topology structure with the incidence relation between the routing addresses and the physical links.

Drawings

Fig. 1 is a flowchart of a backbone network topology discovery method according to an embodiment of the present application;

fig. 2 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

fig. 3 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

fig. 4 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

fig. 5 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

fig. 6 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

fig. 7 is a flowchart of a backbone network topology discovery method according to another embodiment of the present application;

FIG. 8 is a schematic diagram of a backbone network topology network according to the present application;

fig. 9 is a block diagram of a backbone network topology discovery apparatus according to an embodiment of the present application;

fig. 10 is a block diagram of a control apparatus according to an embodiment of the present application.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, embodiments accompanying the present application are described in detail below.

The existing topology discovery method can only obtain the ip address of the router on one path, and the topology discovery can determine each ip address of the router, but cannot determine the possible association relationship between the ip address of the router and the physical link. Thus, when an application such as a fault needs to be performed according to a network topology, accurate positioning cannot be performed due to lack of association with a physical link.

In order to solve the above problem, the present application provides a solution, which determines an association relationship between ip addresses of some routers and physical links by combining adjacency relationships of network devices and characteristics of the physical links.

It should be noted that there are three kinds of existing domestic operator networks, and the topology discovery scheme of the backbone network in the present application is performed in a single operator network, and the topology structure of the backbone network of the operator is discovered; if the topology of the backbone network of other operators needs to be determined, the determination needs to be carried out in the network of other operators.

It should be noted that the operator network in the present application is not limited to a domestic operator network, and may also be a foreign operator network, which is not limited in this application. In addition, if the operator network is replaced by the global internet, the possible topology of the backbone network to obtain the global internet can be finely adjusted based on the present application, and the present application is only described by taking the domestic operator as an example, and the remaining extensions are not limited herein.

For ease of understanding, the following terms that may be used are explained herein:

RTT, round trip delay, is an important performance indicator in computer networks, and represents the total time delay experienced from the time when a sender sends data to the time when the sender receives an acknowledgement from a receiver (the receiver sends an acknowledgement immediately after receiving the data).

And routing hop count, namely, one hop is the number of hops when a data packet passes through one router, and the number of the routers passing through is the hop count. The total hop number from A to B is more than or equal to the hop number of any transfer route from A to AB.

The Transmission Control Protocol (TCP) is a connection-oriented, reliable transport layer communication Protocol based on a byte stream.

UDP, User data packet Protocol (User Datagram Protocol), which provides a method for applications to send encapsulated IP packets without establishing a connection.

ICMP Internet Control Message Protocol (Internet Control Message Protocol). It is a subprotocol of the TCP/IP protocol cluster for passing control messages between IP hosts and routers. Control messages refer to messages of the network itself, such as network traffic down, whether a host is reachable, whether routing is available, etc.

The backbone network comprises a national backbone network layer and a provincial backbone network layer, wherein the provincial backbone network layer is applied to network connection between provincial cities, and the two provincial backbone networks are connected through the national backbone network; when a city in one province needs to perform network communication with a city in another province, the sent data is sent to the node of the corresponding backbone network layer through the province backbone network, is transmitted to the node of the corresponding backbone network layer in the other province through the node, and then is transmitted to the province backbone network of the other province and is transmitted to the corresponding city. The nodes transmitted to the corresponding national backbone network layer in the province are called as the inbound backbone, and the nodes transmitted to the corresponding national backbone network layer in another province are transmitted to the backbone network in another province as the outbound backbone.

The backbone network layer comprises backbone core nodes and backbone access nodes, and the backbone access nodes transmit data to another backbone access node through the backbone core nodes when transmitting the data; therefore, the inbound backbone node and the outbound backbone node correspond to a backbone access node or a backbone core node in a backbone network layer.

In addition, the country has a plurality of provinces, prefectural cities and autonomous regions, which are all equal to the provinces in a specific backbone network, so the provinces in the application refer to the provinces, the prefectural cities or the autonomous regions without special description.

The embodiment of the application provides a backbone network topology discovery method, which can be executed by a backbone network topology discovery device, wherein the backbone network topology discovery device can be integrated in electronic equipment such as a computer, a server, a computer, a server cluster and a data center. Fig. 1 is a flowchart illustrating a topology discovery method for a backbone network according to an embodiment of the present application; the backbone network topology discovery method comprises the following steps:

s100, acquiring detection data between areas capable of reflecting the adjacency relation of network equipment on a backbone network, wherein the detection data at least comprises the detection data between the areas meeting preset conditions;

s200, network topology discovery is carried out according to the detection data so as to obtain the network topology of the backbone network.

The area division is based on the detection data, and the area which can acquire the detection data is the correctly divided area; for example, if the detection device is set as a detection node to acquire the detection data, one detection device may be divided into one area, or a plurality of detection devices may be divided into one area, but a location where no detection device is set may not be divided into one area. However, it should be noted that, due to the characteristics of the backbone network, information management is performed in units of different provinces, and therefore, the division of the areas should not be across provinces, that is, one province may include a plurality of areas, but one area should not include two or more than two provinces or should not include detection devices deployed in two or more than two provinces.

For example, on the premise that the detection data (with the detection device deployed) can be acquired, a building in the hai lake area of beijing city can be divided into one area, a part of the south of the river except for zheng city can be divided into one area, and a aster area and a gaoyang county of the baoding city of north of the river can be divided into one area.

The routing addresses at two ends of the physical link have special characteristics, for example, the last bit phase difference of the IP address is 1, so that the sequential routing addresses from the A area to the B area and from the B area to the A area are obtained, the routing addresses at two ends of at least one physical link can be determined according to the characteristics, and the possible association relationship between the IP address of the router and the physical link is established.

Therefore, by acquiring the probe data reflecting the adjacency relation of the network equipment on the backbone network, the incidence relation between part of the routing addresses on the backbone network and the physical links can be determined, and the network topology structure with the incidence relation between the routing addresses and the physical links is constructed.

It should be noted that the topology of the backbone network in the operator network is similar to the backbone network assumed in this application, but may still be partially different, for example, in the actual topology, adjacent provinces a and B are divided into the same intra-provincial backbone network, in this case, the probe data of provinces a and B cannot be regarded as the probe data between the inter-provincial regions. The embodiment of the present application provides another backbone network topology discovery method, which is similar to the backbone network topology discovery method described above, and the difference is that in this embodiment, a manner of acquiring probe data is specifically described.

The S100, acquiring the detection data between the areas on the backbone network, includes:

Because the detection nodes are deployed in the region, the acquired detection data can be stored in the detection nodes in advance and sent to the statistical equipment when needed; the detection node may be any device disposed in a network of a corresponding area, such as a computer, a server for cloud computing, and the like. In the application, the detection node has an ip address and can send a detection packet, and the other functions are not limited.

And sending the data to the statistical device when needed, wherein in one embodiment, a sending time point can be preset, and each detection node sends the detection data to the statistical device at the time point uniformly. In another embodiment, the statistical device sends a data acquisition request to each probe node when needed, and the probe node sends probe data after receiving the data acquisition request. In another implementation, similar to the previous embodiment, the difference is that when the statistical device needs the probe data, it first determines which areas of the probe data are needed, and then sends a data acquisition request to some of all the probe nodes that are needed.

As to the acquisition mode of the probe data, in an embodiment, the probe data of the probe node sends a probe packet to some or all probe nodes deployed in other areas through the probe node, and receives data returned by a link transit route based on the probe packet to acquire the data. In another embodiment, the probe data of the probe node is obtained by the probe node sending a probe packet set to some or all probe nodes deployed in other areas and receiving a return packet returned by a link transit route based on the probe packet.

For example, a probe node of the a-zone establishes a communication link to a probe node of the B-zone, the communication link including sequentially arranged transit routes from the a-zone to the B-zone; the detection node in the area A sends a detection packet, and the detection packet is transmitted to the corresponding detection node in the area B through a transfer route which is arranged in sequence. The transit route returns route address data to the detection node in the area A after the detection packet path, and the detection node calculates the round-trip delay (rtt) corresponding to the route address data according to the received route address data, the received time and the time of sending the detection packet.

For example, a probe node of the a-zone establishes a communication link to a probe node of the B-zone, the communication link including sequentially arranged transit routes from the a-zone to the B-zone; the detection node in the area A sends detection packets to the sequentially arranged transit routes, and also sends detection packets to the corresponding detection node in the area B, the transit route and the detection node in the area B return a return packet after receiving the detection packets, and the detection node in the area A calculates the corresponding round-trip delay according to the information in the return packet and the return time of the return packet.

Since the number of the routing nodes specifically included in the backbone network is extremely large, in order to perform topology discovery on the backbone network, it is necessary to cover the routing nodes in the backbone network as much as possible in the probe data, that is, when link connections are established between the probe nodes, more routing nodes in the backbone network need to be passed through as much as possible.

For the ICMP protocol, links between probing nodes based on the ICMP protocol are relatively fixed, and routing nodes passed by each probing are also relatively fixed, so that the coverage area is very small. In this case, to complete the coverage in a wide range, a very large number of probe nodes need to be arranged in each area, and the cost is too high.

For the TCP protocol and the UDP protocol, a port is used for providing service for communication, and under the condition of port determination, a link for detecting the quality of a node based on the TCP protocol or the UDP protocol is relatively fixed, but after the port is changed, the link for detecting the quality of the node based on the TCP protocol or the UDP protocol is changed. Therefore, when the port is changed during each detection, the passing routing node is not fixed, and the large-scale coverage can be completed through multiple detections.

In one embodiment, in the acquisition of the probe data of the probe node, the probe is performed based on a TCP protocol and/or a UDP protocol, and a port of the probe node that sends a probe packet is randomly selected at each time of the probe. In another embodiment, the probing is based on the TCP protocol and any network layer or transport layer protocol other than the UDP protocol. In another embodiment, the probing is based on the UDP protocol and any network layer or transport layer protocol other than the TCP protocol.

On one hand, the maximum routing coverage is obtained through the characteristics of high randomness and wide coverage of the TCP/UDP protocol, on the other hand, the characteristic of high stability of any network layer or transport layer protocol is combined with the content of the TCP/UDP protocol for judgment, so that the large data fluctuation caused by the TCP/UDP protocol is reduced, and the judgment accuracy is improved.

In an embodiment, in the acquisition of the probe data of the probe node, any network layer or transport layer protocol is an ICMP protocol or a DCCP protocol.

In one embodiment, the probing nodes probe each other, and according to symmetric probing data obtained by probing each other, association between a possible physical link and routing address information can be determined.

In one embodiment, the probing nodes perform parallel probing based on a transport protocol, and multiple transport protocols simultaneously transmit probe packets. In another embodiment, multiple transport protocols send probe packets in sequence. The total detection time of one detection node to other detection nodes is shortened through parallel detection.

In an embodiment, when the probe node performs probing, low frequency probing may be performed, probe data may be gradually accumulated, and probe data within a certain time range may be sent as needed.

For example, the detection node performs detection on the other detection nodes once every 30 minutes, and a source port is randomly selected during detection; the method comprises the following steps that a plurality of transit routes pass through in one detection, each transit route returns a return packet, and the return packet contains route address information; the destination detection node also returns routing address information and packet loss information.

In one embodiment, at least two detection nodes are arranged in each area, and the detection nodes in the other areas are subjected to low-frequency uninterrupted detection.

An embodiment of the present application provides another backbone network topology discovery method, which is similar to the backbone network topology discovery method described above, and the difference is that, as shown in fig. 2, the S200 performs network topology discovery according to the probe data to obtain a network topology of the backbone network, including:

s210, determining the outbound trunk information and the inbound trunk information based on the one-way detection data among the regions;

and S230, constructing the network topology of the backbone network according to the one-way detection data among the areas and the corresponding outbound trunk information and inbound trunk information.

Unidirectional detection data between the areas are all detection data from all detection nodes of the first area to all detection nodes of the second area; according to the one-way detection data between the areas, the foreign trunk information and the inbound trunk information from the first area to the second area can be determined.

In this embodiment, if the unidirectional probe data between the regions is probe data between trans-provincial regions, the information of the domestic trunk and the information of the domestic trunk from the first region to the second region may be determined, and if the unidirectional probe data between the regions is probe data between non-trans-provincial regions, the information of the domestic trunk and the information of the domestic trunk from the first region to the second region are both null values.

The one-way detection data between the two regions have a common characteristic, the detection data is summarized on the basis of the one-way detection between the two regions, and the foreign trunk information and the domestic trunk information can be determined according to the common characteristic.

For example, the a-zone includes probe nodes a1, a2, and the B-zone includes probe nodes B1, B2; the probe data of A1 to B1 are one-way probe data, and for convenience of recording, are recorded as A1B1 probe data; similarly, the B1A1 probe data is the inverse of the A1B1 probe data; the unidirectional probe data between the regions from region a to region B, which is referred to herein as AB city probe data, is obtained by summarizing the probe data from region A1B1, probe data from region A1B2, probe data from region A2B1, and probe data from region A2B 2.

In one embodiment, the probe data includes routing address information and round trip delay for each hop of a plurality of routing links; as shown in fig. 3, the step S210 of determining the information of the foreign trunk and the information of the inbound trunk based on the one-way probe data between each of the regions includes:

s211, splitting the unidirectional detection data among the regions based on the hop count, and determining the routing address information set and the round-trip delay of each hop count;

and S213, determining the hop count and the routing address data of the outbound trunk and the hop count and the routing address data of the inbound trunk according to the routing address information set and the round-trip delay of each hop count.

The unidirectional detection data between each region comprises a plurality of routing links, and each routing link comprises routing address information and round-trip delay of each hop of route; splitting is carried out based on the hop count, routing address information and corresponding round-trip delay of all ith hops can be obtained, statistics is carried out on the routing address information and the corresponding round-trip delay of each hop, all the routing address information of the hop count can be obtained and used as a routing address information set, and the round-trip delay of the hop count can also be determined according to the round-trip delay corresponding to the routing address information.

Determining a routing address information set and a round trip delay of each hop count, in an embodiment, the method specifically includes: extracting a routing address information set with a certain hop count and the round-trip delay corresponding to each routing address information, merging the same routing address information, determining all round-trip delays corresponding to the routing address information, and taking the median of all round-trip delays as the round-trip delay corresponding to the routing address information; and taking the occurrence frequency of each address information in the routing address information set as a weight basis, and obtaining the round-trip delay of the hop number by weighted average of the round-trip delay.

Determining a routing address information set and a round trip delay of each hop count, in another embodiment, specifically: extracting a routing address information set of a certain hop count and the round-trip delay corresponding to each routing address information, firstly, screening effective round-trip delay, and then determining the round-trip delay of the hop count according to the previous embodiment through the screened round-trip delay and the corresponding routing address information.

In the above embodiment, after determining all round-trip delays corresponding to the routing address information, the round-trip delay corresponding to the routing address information may be determined according to the median of all round-trip delays, may also be determined according to the mean, variance, quantile, and the like of all round-trip delays, and may also be determined comprehensively according to the characteristics of the mean, variance, quantile, and the like added to the round-trip delays on the basis of the median of the round-trip delays; in the specific implementation process, the implementation effect of the median is the best.

When the detection node detects, a detection packet is sent to a routing address of each hop on a link, if a return packet of the routing address is received and the round-trip delay is within a certain range of the round-trip delay at the tail end of the link, the round-trip delay can be considered to be effective, and all the round-trip delays can be screened on the basis of the round-trip delay, so that the effective round-trip delay can be screened out. If the probing node does not receive the return packet of the routing address, the corresponding round-trip delay may be set to a null value or a very high default value.

According to the routing address information set and the round-trip delay of each hop count, determining the hop count and the routing address data of the foreign trunk and the hop count and the routing address data of the domestic trunk, wherein in one implementation mode, for the routing address information set and the round-trip delay of each hop count in one direction between each region, the hop count and the routing address data of the domestic trunk are determined in the following mode: and calculating the difference (the difference is an absolute value) between the round trip delay of the second hop and the round trip delay of the next hop from the second hop, and if the difference is greater than 1/p of the round trip delay of the last hop, the hop count is the number of the domestic trunk hops, otherwise, the next hop is judged.

Similarly, the hop count and routing address data of the outbound trunk and the hop count and routing address data of the inbound trunk are determined according to the routing address information set and the round trip delay of each hop count, and in one embodiment, for the routing address information set and the round trip delay of each hop count between each area in a single direction, the hop count and routing address data of the outbound trunk are determined in a manner that: taking the difference (the difference is an absolute value) between the round-trip delay of each hop (except the last hop) and the round-trip delay of the last hop as the characteristic value of each hop, starting from the last hop, calculating the difference (the difference is an absolute value) between the characteristic value of the calculator and the characteristic value of the previous hop, and if the difference is more than 1/q of the round-trip delay of the last hop, the hop count is the number of the inbound trunk hops, otherwise, the previous hop is judged.

For example, the round trip delay of each hop count of the area AB is 1, 2, 2.5, 4, 5, 7, 8, 9 in turn; the hop count of the inbound trunk is determined by calculating the difference between the round-trip delay of the 2 nd hop and the round-trip delay of the next hop to be 0.5 from the second hop, and judging that the difference is not more than 1/p of the round-trip delay of the last hop, namely 9/p; and continuously judging the 3 rd hop, wherein the difference value of the round trip delay of the 3 rd hop and the round trip delay of the next hop is 1.5 and is more than 9/p, and the third hop is one hop of the national trunk. The hop count determination mode of the foreign trunk is that the difference value of the round trip delay of each hop count and the round trip delay of the last hop is calculated firstly and is used as the characteristic value of each hop count, and the characteristic value of each hop count of the calculated area AB is 8, 7, 6.5, 5, 4, 2, 1 and 0 in sequence; calculating the difference value of the characteristic value of the 2 nd hop from the last hop to be 1 from the last hop, and judging that the difference value is not more than 1/q, namely 9/q, of the round-trip delay of the last hop; and continuously judging the last 3 rd hop, wherein the difference value between the characteristic value of the last hop and the characteristic value of the last hop is 2, and if the difference value is more than 9/q, the last 3 rd hop is a hop of the foreign trunk.

It should be emphasized that the above example is only an illustrative example, and the round-trip delay of each hop count of the area AB in the real example is not obtained by comprehensive processing of multiple actual data, and is not raw data (the raw data obtained by detection is unstable, and a more stable round-trip delay needs to be obtained by comprehensive processing); the change of the domestic trunk and the foreign trunk is large, and the specific possible values are 1, 1.5, 2, 6, 6.5 and 7;

in one embodiment, the threshold parameter p and the threshold parameter q are determined by a heuristic search algorithm and are fixed values; in another embodiment, the threshold parameter p and the threshold parameter q are determined according to the statistic of the historical data, such as standard deviation, quantile number, etc., and after the historical data is updated, the threshold parameter p and the threshold parameter q determined by the statistic are also updated. In another embodiment, the values of the threshold parameter p and the threshold parameter q corresponding to different regions are different.

In an embodiment, the S211, after splitting the unidirectional probe data between each region based on the hop count and determining the routing address information set and the round trip delay of each hop count, further includes:

s212, smoothing is performed based on the round trip delay per hop count.

Performing smoothing processing based on the round trip delay of each hop count, which in an embodiment specifically is: and smoothing the round-trip delay of each hop number by a preset algorithm, and setting the round-trip delay judged to be invalid as a null value. In another embodiment, the round trip delay per hop is smoothed based on a moving average. In another embodiment, the round trip delay per hop is smoothed by means of a filter.

Performing smoothing processing based on the round trip delay of each hop count, wherein in another embodiment, the smoothing processing specifically includes: judging from the 2 nd hop in sequence, if the effective data proportion of the ith hop is small, the effective data proportion of the (i-1) th hop is large, and the round-trip delay of the (i-1) th hop is higher than the round-trip delay of the ith hop by a certain threshold value x, judging that the round-trip delay of the ith hop is invalid; and if the effective data proportion of the ith hop is small, the effective data proportion of the (i + 1) th hop is large, and the delay of the (i + 1) th hop is lower than the delay of the ith hop by a certain threshold value y, judging the round-trip delay of the ith hop to be invalid. The size of the effective data ratio can be judged by a preset threshold value.

In one embodiment, the threshold value x and the threshold value y are determined by a heuristic search algorithm and are fixed values; in another embodiment, the threshold value x and the threshold value y are determined according to the statistic of the historical data, such as standard deviation, quantile number, etc., and after the historical data is updated, the threshold value x and the threshold value y determined by the statistic are also updated. In another embodiment, the values of the threshold x and the threshold y are different between different regions.

For each hop, the effective data proportion of the round trip delay of the hop should be kept in a proper range, and if the effective data proportion of the round trip delay of adjacent hops is greatly different, one of the effective data proportion may be invalid data; in addition, as the number of hops increases, the round-trip delay should also increase accordingly, and if the round-trip delay is rather reduced, one of the round-trip delay may be invalid data.

By integrating the effective data ratio of the round trip delay and the difference value of the adjacent round trip delay, the effectiveness of the round trip delay can be accurately judged.

For example, the ratio of valid data of the 3 rd hop round trip delay is less than 20%, and the ratio of valid data of the 2 nd hop round trip delay is more than 80%; or the difference between the effective data occupation ratio of the 3 rd hop round-trip delay and the effective data occupation ratio of the 2 nd hop round-trip delay is more than 55%, the data is considered to have abnormality in the effective data occupation ratio of the round-trip delay; if the round-trip delay of the 2 nd hop is higher than the round-trip delay of the 3 rd hop by a preset threshold value x, the round-trip delay is considered to be abnormal in the aspect of increasing the round-trip delay sequence; if there is an anomaly in both aspects, the 3 rd round trip delay is determined to be invalid.

The embodiment of the present application provides another backbone network topology discovery method, which is similar to the backbone network topology discovery method described above, and is different in that, as shown in fig. 4, the unidirectional probe data between the regions is probe data from a source region to a destination region;

the S210, after determining the information of the foreign trunk and the information of the domestic trunk based on the unidirectional probe data between each region, further includes:

s220, the foreign trunk information and the domestic trunk information between the areas are corrected based on the aggregation characteristics of the same source area and the same destination area.

The same source region detects a plurality of destination regions, or a plurality of source regions detect the same destination region, and the routing address information can be embodied as scattered or converged. For the detection of the homologous regions, a group of regions can enter the country trunk at the route with the positive number and the same hop number; for the same-purpose detection, a group of regions can go out of the country in the route with the inverse number of hops, which is the aggregation characteristic of the same source region and the same destination region.

In one embodiment, as shown in fig. 5, S220, modifying the outbound trunk information and the inbound trunk information between the domains based on the aggregation characteristics of the same source domain and the same destination domain, includes:

s221, correcting the inbound trunk information among the areas corresponding to the source area based on the inbound trunk information of the same source area;

s222, the information of the foreign stems between the areas corresponding to the target area is corrected based on the information of the foreign stems of the same target area.

As shown in fig. 6, the S221, based on the trunk information of the same source region, corrects the trunk information between the regions corresponding to the source region, and includes:

s2211, counting the inbound trunk information of the same source region, and sorting according to the occurrence frequency of the hop count of the inbound trunk;

s2212, determining a union set of the sorted domestic trunk information based on the routing address information set of the domestic trunk information corresponding to each sort;

and S2213, according to the hop count and the routing address information set between the non-top-ranked areas, correcting the trunk information between the areas based on the union and the hop count ranked in advance.

Correcting the inbound trunk information between the regions based on the union and the hop count sorted in advance, and in one embodiment, the method specifically comprises the following steps: extracting a first sorted union set and hop counts, judging whether the union set belongs to the first sorted union set according to a routing address information set with the same hop count among the areas, and modifying the inbound trunk information among the areas into the first sorted hop counts if the union set belongs to the first sorted union set; if not, extracting a second sorted union set and hop count, judging whether the union set belongs to the second sorted union set according to a routing address information set of the same hop count among the regions, and if so, modifying the inbound trunk information among the regions into the second sorted hop count; if not, continuing to extract and judge; if the online sorting extraction is finished, judging that the results do not belong to the same, maintaining the original domestic trunk hop count, and not modifying.

For example, in the inter-domain inbound trunk information, the hop counts of the inbound trunks of the domains AB, AC, AD, AE, AF, AG, AH, AI, AJ, AK, AL, and AM are 5, 6, 5, 7, 4, 7, 6, 5, and 6, respectively, and are ranked 555566677744 according to the frequency of occurrence of the hop counts of the inbound trunks (hop count 6 and hop count 7 are the same in frequency, and may be randomly ranked, this time, 6 top row), and the corresponding inter-domain is AB, AD, AK, and AL; AC. AJ, AM; AE. AH and AI; AF. And AG. A routing address information set of the inbound trunk information corresponding to the first 5555 in frequency ordering determines a first ordered union, namely a union of routing address information sets of 5 th hops of AB, AD, AK and AL; similarly, the union of the routing address information sets of the 6 th hop of the second 666, i.e., the AC, AJ, AM, the union of the routing address information sets of the 7 th hop of the third 777, and the AE, AH, AI are determined (the last ordered union may not be computed). Then respectively correcting the inbound trunk information of the combination of the other areas except the first sequence, taking AM as an example and the hop count of the first sequence is 5, firstly determining the routing address information set of the 5 th hop of the AM, judging whether the routing address information set belongs to the union set of the first sequence, and if the routing address information set belongs to the union set of the first sequence, correcting the inbound trunk information of the AM to the 5 th hop; if not, checking the second ordering, wherein the hop count is 6, firstly determining the routing address information set of the 6 th hop of the AM, judging whether the routing address information set belongs to the union set of the second ordering, and if so, correcting the inbound trunk information of the AM to the 6 th hop; if not, the 7 th hop of the inbound trunk information is still maintained, and no correction is performed.

Similarly, the information of the foreign stems between the areas corresponding to the same destination area may be corrected based on the information of the foreign stems of the destination area.

The embodiment of the present application provides another backbone network topology discovery method, which is similar to the backbone network topology discovery method described above, and the difference is that, as shown in fig. 7, in S230, a network topology of the backbone network is constructed according to inter-region unidirectional probe data and corresponding outbound trunk information and inbound trunk information, where the method includes:

s231, according to the outbound trunk information and the inbound trunk information among the regions, dividing the one-way detection data among the regions into a source region province internal network, a target region province internal network and a national trunk network;

s232, constructing a network topology of the backbone network according to the divided unidirectional detection data among the regions;

and S233, based on the routing address characteristics of the physical link, determining the routing address information of the two ends of the physical link and the routing address information of the national trunk switch group according to the unidirectional detection data and the reverse detection data between the regions, and correcting the network topology of the backbone network.

The one-way detection data between the areas are divided into a source area provincial network, a target area provincial network and a national network, and in one implementation mode, the method specifically comprises the following steps: finding the inbound trunk hop count and the outbound trunk hop count in the link routing data of the detection data, wherein the routing data before the inbound trunk hop count in the link routing data is the provincial backbone network where the source region is located; after the domestic trunk hop count, the routing data before the domestic trunk hop count is the domestic trunk network data; the routing data after the foreign trunk hop count is the provincial backbone network where the target area is located; based on the classification, the routing address information in the detection data is used as nodes, and the adjacent relation in the link routing is used as the adjacent relation between the nodes to construct the network topology of the backbone network.

The routing address characteristic of the physical link is that the difference of the last digit of the routing ip addresses at two ends of one physical link (optical cable) is 1.

Based on the routing address characteristics of the physical link, according to the unidirectional detection data and the reverse detection data between the regions, determining routing address information of two ends of the physical link and routing address information of the national trunk switch unit, in an implementation manner, specifically: if the difference between the last digit of two routing addresses between the areas is 1, and one of the two routing addresses belongs to unidirectional detection data and the other one belongs to reverse detection data, the two routing addresses can be determined to be two ends of a physical link; the inbound trunk routing address information of the unidirectional probe data and the outbound trunk routing address information of the reverse probe data between the two areas can be determined to belong to the same group of switches; similarly, the outbound trunk routing address information of the unidirectional probe data and the inbound trunk routing address information of the reverse probe data between the two areas, such as the inbound trunk routing address information, may also be determined to belong to the same group of switches.

For example, as shown in fig. 8, the square boxes in the figure represent two trunk switches, the circular dots represent IP addresses (routing address information), and the two IP addresses on the left side of the left square box are the inbound trunk IP addresses from area a to area B; the two IP addresses on the left side of the right square box are the outbound trunk IP addresses from the area A to the area B; the two right IP addresses of the right square frame are the inbound trunk IP addresses from the area B to the area A; the two IP addresses on the right side of the left square frame are the outbound trunk IP addresses from the area B to the area A; and the outbound trunk IP addresses (two IP addresses on the left side of the right square box) from the area a to the area B and the outbound trunk IP addresses (two IP addresses on the right side of the left square box) from the area B to the area a are respectively connected through a physical link (optical cable), so that the difference between the last digit of the corresponding IP addresses is 1.

On the contrary, under the condition of determining the one-way and reverse outbound and inbound trunk IP addresses between the area A and the area B, the corresponding trunk switch set can be determined, but whether the trunk switch set is a single trunk switch or not cannot be determined; after determining the IP address of the trunk switch group, the connection relationship of the IP address through the physical link (optical cable) can be determined (if there is a direct connection relationship).

In an embodiment, the S230, constructing a network topology of the backbone network according to the inter-region unidirectional probe data and the corresponding outbound trunk information and inbound trunk information, specifically further includes: determining a target area passing through the same inbound trunk exchange unit based on the inbound trunk information of the same source area; determining a source region passing through the same outbound trunk exchange unit based on the outbound trunk information of the same target region; and modifying the network topology of the backbone network.

For the inbound trunk information based on the same source region, the destination region via the same inbound trunk switch group is determined, which is illustrated as follows: the inbound trunk IP address sets of the areas AB, AC and AD of the same source area are { a, b, c }, { d, e and f }, respectively, wherein the inbound trunk IP address sets of the area AB and the area AC are the same, and the area AB and the area AC communicate through the same set of inbound trunk switch sets. Similarly, a source zone via the same foreign trunk switch group is determined based on the foreign trunk information of the same destination zone.

The number of the existing inbound trunk IP addresses and outbound trunk IP addresses between two areas is generally kept within 10, and the IP addresses with the number can be generally completed by arranging a switch or a group of switches. It should be noted that, because the usage of the current trunk switch, the inbound trunk IP addresses of the same source area and different destination areas are only identical and completely different comparison results, and there is no case of partial identity, the usage of the same trunk switch (group) can be directly confirmed.

And in the network topology of the backbone network, marking or replacing the confirmed use condition of the national trunk exchange unit to finish the correction.

The topology discovery method of the backbone network uses the detection nodes in each area to perform parallel and low-frequency detection based on UDP, TCP and other multi-protocols, and can comprehensively cover the routing nodes passing between the two areas with low cost by randomizing the source ports.

According to the backbone network topology discovery method, a network construction party does not need to provide any prior knowledge, and a routing address information set of each hop route between two areas, the geographic position of the routing address (divided into a specific provincial backbone network or a national backbone network) and the incidence relation of the routing address on a physical link can be determined according to the round-trip delay and the routing address.

When the method is used specifically, the constructed backbone network topology structures of three operators reach 95% of accuracy rate through the backbone network topology discovery method. And the construction of the backbone network topology from the probe data to the latest version can be completed in one quarter of a second.

In an implementation mode, the determined hop count, the route address set and the round-trip time delay between the areas are determined by detecting data between the areas, and the determined hop count, the route address set and the round-trip time delay between every two of the areas can be determined simultaneously in a parallel mode, so that the calculation time is greatly reduced.

In another embodiment, when the outbound trunk information and the inbound trunk information between the areas are corrected based on the aggregation characteristics of the same source area and the same target area, the outbound trunk information and the inbound trunk information of a plurality of same source areas and/or the same target area can be corrected simultaneously in a parallel manner, so that the correction time is greatly reduced.

In another embodiment of the backbone network topology discovery method, the area is a city.

Therefore, on one hand, the detection equipment is arranged by taking the city as a unit, a plurality of (such as 3-5) detection equipment are arranged to meet the basic detection requirement, and compared with a division mode of a smaller area, the method can save a great amount of detection equipment, and on the other hand, in the constructed backbone network topological graph, the method is more suitable for the division of the existing geographic position by taking the city as a unit, and is more convenient and more suitable in the actual use process.

In another embodiment, when constructing the network topology of the backbone network, the method for discovering the topology of the backbone network can also screen the unidirectional detection data between two areas and the corresponding outbound trunk information and inbound trunk information and determine the routing address information at the two ends of the physical link and the routing address information of the national trunk exchange unit in a multi-task parallel mode so as to greatly reduce the construction time of the network topology of the backbone network.

The embodiment of the present application provides a backbone network topology discovery device, which is used for executing the backbone network topology discovery method described in the foregoing content of the present application, and the backbone network topology discovery device is described in detail below.

As shown in fig. 9, the backbone network topology discovery apparatus includes:

a data obtaining module 101, configured to obtain probe data between areas that can represent an adjacency relation of network devices on a backbone network, where the probe data at least includes probe data between areas that meet a preset condition;

a topology discovery module 102, configured to perform network topology discovery according to the probe data to obtain a network topology of the backbone network.

The regions meeting the preset conditions are trans-provincial regions.

The area is a city.

The topology discovery module 102 includes:

a trunk information determination unit for determining the outbound trunk information and the inbound trunk information based on the one-way probe data between each region;

and the network topology construction unit is used for constructing the network topology of the backbone network according to the one-way detection data among the areas and the corresponding outbound trunk information and inbound trunk information.

The trunk information determination unit includes:

a hop count splitting subunit, configured to split the unidirectional probe data between each region based on hop counts, and determine a routing address information set and a round-trip delay for each hop count;

and the hop count determining subunit is used for determining the hop count and the routing address data of the outbound trunk and the hop count and the routing address data of the inbound trunk according to the routing address information set and the round-trip delay of each hop count.

After the operation of the hop count splitting subunit is performed, the trunk information determining unit further includes:

and a smoothing subunit, configured to perform smoothing based on the round trip delay per hop count.

After performing the operation of the trunk information determining unit, the topology discovery module 102 further includes:

and a trunk correction unit for correcting the outbound trunk information and the inbound trunk information between the regions based on the aggregation characteristics of the same source region and the same destination region.

The national stem correction unit includes:

a country trunk correction subunit, configured to correct, based on the country trunk information of the same source region, the country trunk information between the regions corresponding to the source region;

and a stem of departure correction subunit configured to correct stem of departure information between areas corresponding to the target area, based on stem of departure information of the same target area.

The inbound trunk correction subunit is specifically configured to: counting the inbound trunk information of the same source region, and sequencing according to the occurrence frequency of the hop count of the inbound trunk; determining a union set of the sequenced inbound trunk information based on the routing address information set of the inbound trunk information corresponding to each sequence; and according to the hop count and the routing address information set between the non-head-ranked areas, correcting the trunk information between the areas based on the union and the hop count ranked in advance.

The network topology constructing unit includes:

the network dividing unit is used for dividing the one-way detection data among the regions into a source region provincial network, a target region provincial network and a national trunk network according to the domestic trunk information and the domestic trunk information among the regions;

a topology construction subunit, configured to construct a network topology of the backbone network according to the partitioned unidirectional probe data between the regions;

and the topology correction subunit is used for determining routing address information of two ends of the physical link and routing address information of the national trunk switch group according to the unidirectional detection data and the reverse detection data among the areas based on the routing address characteristics of the physical link, and correcting the network topology of the backbone network.

The network topology constructing unit further includes:

a inbound trunk determining subunit, configured to determine, based on inbound trunk information of the same source region, a destination region via the same inbound trunk switch group;

a foreign trunk determination subunit, configured to determine a source region via the same foreign trunk switch group based on the foreign trunk information of the same destination region;

a network modification subunit for modifying a network topology of the backbone network.

The data acquisition module 101 is specifically configured to: sending a detection data acquisition request to part or all of detection nodes deployed in an area, and receiving the detection data returned by the detection nodes according to the detection data acquisition request; or receiving the detection data sent by part or all of the detection nodes deployed in the area at a preset time point.

In the operation execution process of the data acquisition module 101, the probe data of the probe node sends a probe packet to some or all probe nodes deployed in other areas through the probe node, and receives data returned by a link transit route based on the probe packet to acquire the data.

In the operation execution process of the data acquisition module 101, during acquisition of probe data of the probe node, the probe node sends a probe packet to some or all probe nodes deployed in other areas based on at least one of a TCP protocol and a UDP protocol.

The backbone network topology discovery device provided by the above embodiment of the present application and the backbone network topology discovery method provided by the embodiment of the present application have the same inventive concept and have the same beneficial effects as methods adopted, operated or implemented by application programs stored in the backbone network topology discovery device.

Having described the internal functions and structure of the network topology discovery apparatus as shown in fig. 10, in practice, the network topology discovery apparatus can be implemented as a control device including: a communication component 305, a memory 301, and a processor 303.

A memory 301, which may be configured to store a program.

In addition, the memory 301 may also be configured to store other various data to support operations on the control device. Examples of such data include instructions for any application or method operating on the control device, contact data, phonebook data, messages, pictures, videos, and the like.

The memory 301 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The communication component 305 is configured to facilitate wired or wireless communication between the control device and other devices. The control device may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 305 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 305 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

Specifically, the communication component 305 is configured to obtain probe data between areas that can represent an adjacency relationship of network devices on the backbone network, where the probe data at least includes probe data between areas that meet a preset condition.

A processor 303, coupled to the memory 301 and the communication component 305, for executing programs in the memory 301 for:

In one embodiment, the processor 303 is specifically configured to:

In one embodiment, the probe data includes routing address information and round trip delay for each hop of a plurality of routing links; based on this, the processor 303 is specifically configured to:

In one embodiment, the unidirectional detection data between the regions is detection data from a source region to a destination region; based thereon, the processor is further specifically configured to:

In one embodiment, the processor is specifically configured to:

In one embodiment, the communication component is specifically configured to:

In one embodiment, the area is a city.

In this application, only some components are schematically shown in fig. 10, and it is not meant that the server device includes only the components shown in fig. 10.

The control device provided by this embodiment is based on the same inventive concept as the backbone network topology discovery method provided by this embodiment, and has the same beneficial effects as the method adopted, operated or implemented by the application program stored in the control device.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for discovering a topology of a backbone network, comprising:

2. The method of claim 1, wherein the inter-region satisfying the predetermined condition is a trans-provincial inter-region.

3. The method according to claim 2, wherein the performing network topology discovery according to the probe data to obtain the network topology of the backbone network comprises:

4. The method of claim 3, wherein the probe data comprises routing address information and round trip delay for each hop of the plurality of routing links;

5. The method of claim 4, wherein the splitting of the unidirectional probe data between each of the regions based on the hop count, and after determining the routing address information set and the round trip delay of each hop count, further comprises:

smoothing is performed based on the round trip delay per hop count.

6. The method according to any one of claims 3-5, wherein the uni-directional probe data between the regions is probe data from a source region to a destination region;

7. The method as claimed in claim 6, wherein the modifying the information of the foreign stems and the information of the domestic stems between the regions based on the aggregation characteristics of the same source region and the same destination region comprises:

8. The method as claimed in claim 7, wherein the modifying the inbound trunk information between the regions corresponding to the source region based on the inbound trunk information of the same source region comprises:

9. The method according to any one of claims 3-5, wherein the constructing the network topology of the backbone network according to the one-way probe data between the regions and the corresponding outbound trunk information and inbound trunk information comprises:

10. The method of claim 9, wherein the constructing the network topology of the backbone network according to the inter-region unidirectional probe data and the corresponding outbound trunk information and inbound trunk information further comprises:

and modifying the network topology of the backbone network.

11. The method of any of claims 2-5, wherein the obtaining probe data between regions on the backbone network comprises:

12. The method according to claim 11, wherein the probe data of the probe node is obtained by sending a probe packet to some or all probe nodes deployed in other areas through the probe node and receiving data returned by a link transit route based on the probe packet.

13. The method according to claim 12, wherein in the acquisition of the probe data by the probe node, the probe node transmits a probe packet to some or all of the probe nodes deployed in other areas based on at least one of a TCP protocol and a UDP protocol.

14. The method of any one of claims 1-5, wherein the area is a city.

15. A control apparatus, characterized by comprising: a memory, a processor, and a communications component;

the memory for storing a program;

16. A backbone network topology discovery apparatus, comprising: