WO2011015103A1

WO2011015103A1 - Method for collecting topology information of distributed network and distributed network thereof

Info

Publication number: WO2011015103A1
Application number: PCT/CN2010/075367
Authority: WO
Inventors: 胡永生; 陈志峰
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-08-07
Filing date: 2010-07-21
Publication date: 2011-02-10
Also published as: CN101989995A; CN101989995B

Abstract

Disclosed are a method for collecting topology information of a distributed network and the corresponding distributed network thereof, wherein the method includes: a beginning node in the distributed network sends topology collection request messages to its neighbor nodes. After any node X in the distributed network receives the topology collection request messages, it performs the following processing: node X sends the topology collection request messages to its M neighbor nodes. The M neighbor nodes are the neighbor nodes of node X except the one sending the topology collection request message to node X. After receiving topology collection response messages returned from the M neighbor nodes, or a preset time expires, if at least one received topology collection response message includes network topology information, node X contains the network topology information recorded by the current node and the network topology information in the topology collection response messages received by the current node in the topology collection response message and returns it to one of the nodes sending topology collection request messages to node X itself, wherein M ≥ 1.

Description

Topology information collection method and distributed network for distributed network

Technical field

The present invention relates to the field of communications, and in particular, to a topology information collection method for a distributed network and a corresponding distributed network.

Background technique

P2P (Peer-to-Peer, peer-to-peer, peer-to-peer) technology allows users to connect directly to other users' computers for file sharing and exchange, while P2P is also useful for deep search, distributed computing, and collaborative work. .

A P2P network composed of a large number of computer devices has various topological structural relationships. At present, the mainstream topology structure is divided into the following four types: Centralized Topology, fully distributed unstructured topology ( Decentralized Unstructured Topology), Decentralized Structure Topology, and Partially Decentralized Topology.

Among them, the centralized topology is centered on the central server (directory system), and all nodes (ie Peer) are set around the central node. The network with a centralized topology is simple to maintain and the node discovery efficiency is high. However, the entire network is overly dependent on the central server and is prone to bottlenecks. Therefore, it is only suitable for small networks and is not suitable for large-scale network applications.

There is no central server in the fully distributed unstructured topology structure, and the devices in the network are organized in a random manner. The network using the fully distributed unstructured topology structure can quickly find the destination node, and has strong fault tolerance and can support Complex queries, fuzzy queries, etc.

A fully distributed structured topology is a fully distributed structured network using Distributed Hash Table (DHT). DHT can provide accurate discovery operations, and as long as the destination node exists in the network, it can be discovered.

The semi-distributed structure topology structure combines the advantages of the centralized topology structure and the fully distributed structured topology structure, and selects nodes with strong performance (strong information processing capability) as super nodes, and each super node is responsible for a group. Forwarding of messages between ordinary nodes, while super nodes complete messages in super Forwarding between nodes.

Among the four topologies, except for the centralized topology and a few structured networks that use a one-hop routing algorithm (one-hop routing algorithm is also only applicable to smaller networks), there is one in message routing. The commonality is that each node stores only a part of the network routing table. When any node receives a resource lookup request, it always forwards it to the node that is "closer" to the destination node until the destination node is found.

Figure 1 is a schematic diagram of network topology information provided by each node in a fully distributed structured topology network. Each circle in Figure 1 represents a node in the overlay network, and the dashed line represents the logical relationship between the node and the node. Each node only knows the nodes with logical connections.

As shown in Figure 1, node 1 can only know the existence of four nodes: node 2, node 3, node 4, and node 5; node 6 can know the existence of four nodes: node 4, node 5, node 7, and node 8; The network changes, for example, node 5 exits the overlay network, node 1 and node 6 will immediately know the exit of node 5, and will select a new node replacement node 5 according to the algorithm.

It can be seen from the above that in the entire network, all nodes can only know the topology information of a specific part of the overlay network, and no node can know the topology information of the entire network. Therefore, when collecting the topology information of the entire network, it is usually first to select specific nodes in the overlay network, and after obtaining the topology information from these nodes, the complete network topology information is synthesized. However, when collecting the topology information by the above method, it is necessary to know the topology structure of the network in advance, and select a specific node according to the topology structure of the network to complete the collection of the topology information.

At present, the application of P2P technology is very extensive. The topology and routing algorithms used for different scenarios and requirements are numerous and varied. If you use the above method to collect the entire network topology information, you need to use all types of topology structures and The routing algorithm can only be carried out after research. There is no prior art method for collecting network topology information that can be applied to all topology structures and routing algorithms. Summary of the invention

The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art, and to provide a topology information collecting method and a corresponding distributed network of a distributed network that do not depend on a network topology and a routing algorithm. In order to solve the above technical problem, the present invention provides a topology information collection method for a distributed network, the method comprising:

The originating node in the distributed network sends a topology collection request message to its neighboring nodes; after receiving the topology collection request message, any node X in the distributed network performs the following processing:

The node X sends a topology collection request message to the M neighboring nodes of the node; the M neighbor nodes are nodes other than the node that sends the topology collection request message to the node X among the neighbor nodes of the node X;

After receiving the topology collection response message returned by the M neighboring nodes, or after reaching a preset time, if the received at least one topology collection response message includes network topology information, the node X records the node. The network topology information and the topology information in the topology collection response message received by the local node are included in the topology collection response message, and are returned to one of the nodes to which the topology collection request message is sent;

Where M > 1.

Preferably, the method further includes: if the node X sends the topology collection request message to the node, the node X has no other neighboring nodes, and the node X includes the network topology information of the node in the topology collection response. In the message, return to one of the nodes to which the topology collection request message is sent.

Preferably, the method further includes: if all the topology collection response messages received by the node X do not include the network topology information, the node X includes the network topology information of the node in the topology collection response message. One of the nodes to which the topology collection request message is sent.

Preferably, the node X returns the topology collection response message containing the network topology information to the transmitting node from which the node receives the first topology collection request message.

Preferably, the method further includes: after receiving the topology collection request message, the node X determines whether the message is the received first topology collection request message, and if not, returns to the sending node of the message. A topology collection response message containing network topology information.

Preferably, if the node X has no other neighboring nodes except the node to which the topology collection request message is sent, or the topology topology information is not included in all the topology collection response messages received by the node X, the node The network topology information returned by X contains only the node identifier of node X. Preferably, the method further includes: after receiving the topology collection request message, the node X determines, according to the forwarding restriction identifier included in the message, whether to allow the topology collection request message to be sent to the M neighboring nodes of the node. If not allowed, the node X returns a topology collection response message to the node that sends the topology collection request message; the topology collection response message includes network topology information, where the network topology information includes the M The neighboring nodes are information of nodes to which the topology collection request message has not been forwarded;

After receiving the network topology information including the information that the M neighboring nodes are the nodes to which the topology collection request message has not been forwarded, the originating node sends a topology collection request message to the M neighboring nodes. .

Preferably, the method further comprises: before sending the topology collection request message to its neighboring node, the starting node sets an initial value of the forwarding restriction identifier N to NO, and includes the forwarding restriction identifier in the The topology collection request message is sent to its neighbor node;

After receiving the topology collection request message, any node X in the distributed network determines whether to allow sending a topology to the neighboring node of the node according to whether the N value in the message is less than or equal to N1 or greater than or equal to N2. Collecting the request message; if allowed, subtracting the N value, and then including it in the topology collection request message and sending it to the neighboring node of the node;

Where w<0 JL N0< Nl , or w>0 and N0> N2.

Preferably, the topology collection request message includes a topology information collection identifier for uniquely identifying a current topology information collection operation.

Preferably, the topology information collection identifier includes: an identifier of the start node and a message sequence number.

The present invention also provides a distributed network to which the topology information collecting method of the above distributed network is applied.

In summary, any node in the distributed network can obtain the entire network topology information by using the method provided by the present invention, and the method is applicable to a distributed network using any topology structure and arbitrary routing algorithm. At the same time, because the topology information collection request of the present invention is hierarchically copied and distributed, and the mechanism of response grading and reporting is reported, the local traffic of the network can be effectively controlled without causing a big impact on the network. BRIEF abstract

1 is a schematic diagram of network topology information provided by each node in a fully distributed structured topology network;

2 is a flowchart of a method for collecting topology information of a distributed network according to a first embodiment of the present invention; FIG. 3 is a schematic diagram showing a topology of a network for collecting topology information by using the method of the first embodiment of the present invention;

4 is a message interaction diagram of topology information collection by a starting node in the scenario shown in FIG. 3 according to the first embodiment of the present invention;

FIG. 5 is a message interaction diagram of topology information collection by an intermediate node in the scenario shown in FIG. 3 according to the first embodiment of the present invention; FIG.

6 is a message interaction diagram of topology information collection by a stub node in the scenario shown in FIG. 3 according to the first embodiment of the present invention;

7 is a flow chart of a method for collecting topology information of a distributed network according to a second embodiment of the present invention.

Preferred embodiment of the invention

The core idea of the present invention is that a starting node in a distributed network sends a topology collection request message to its neighboring nodes, and after starting topology information collection, any intermediate node in the distributed network (including the phase of the starting node) After receiving the topology collection request message, the neighboring node forwards the topology collection request message to the neighboring node of the node, and after receiving the returned topology collection response message, the network topology information of the node (ie, the node) The information contained in the routing table and the topology collection response message are included in the topology collection response message and returned to the previous node. Finally, the originating node collects topology information of each node in the network.

The invention will now be described in detail in conjunction with the drawings and embodiments.

First embodiment

2 is a flow chart of a method for collecting topology information of a distributed network according to a first embodiment of the present invention. In the flow shown in FIG. 2, according to the different roles of each node in the distributed network in the topology information collection process, each node is defined as: a start node, an intermediate node, and a stub node; wherein, the start node is initiated The node that collects the topology information is the node that needs to summarize and report the topology information of the local node and the neighboring node. The stub node only needs to report the topology information of the local node, and does not need to collect the topology information of the neighboring node. node. The processing flow of the above three types of nodes in the collection process of the topology information will be described below with reference to FIG. 2 respectively.

The main processing steps of the starting node include:

S201a) The starting node that desires to obtain the network topology information queries the routing table of the local node, sends a topology collection request message to all nodes associated with the routing table (called a neighboring node), and waits for the topology returned by the neighboring node. Collect response messages;

S202a) after receiving the topology collection response message returned by all neighboring nodes, the initiating node combines the topology information included in the message with the topology information included in the routing table of the local node into a complete network topology information. Complete the collection process of this topology information.

The main processing steps of the intermediate node include:

S201b) after receiving the topology collection request message sent by the previous node (possibly the intermediate node or the origin node), in order to avoid the broadcast storm, the intermediate node verifies whether the message is the topology collection request message received for the first time; If the topology collection request message is received for the first time, the intermediate node returns a topology collection failure response message to the previous node, and ends the processing process of the intermediate node; otherwise, the next step is performed;

In order to facilitate each node to know whether the same topology collection request message has been received before, the originating node may generate a unique topology information collection identifier for each topology information collection (for example, the topology information collection identifier includes The identifier of the originating node and the unique message sequence number are inserted into the topology collection request message and sent to the neighboring nodes. Each intermediate node must carry the identifier when forwarding the topology collection request message. After receiving the topology collection request message, each node reads the identifier from the message, and then performs a search in the list of received topology information collection identifiers stored in the node, if the same topology information collection identifier is not found. , indicating that the topology collection request message is the topology collection request message received for the first time. The topology collection failure response message may be a topology collection response message that does not include any topology information.

S202b) The intermediate node queries the routing table of the local node, and sends a topology collection request message to the neighboring node other than the node that has sent the topology collection request message to the intermediate node, and waits for the corresponding topology collection response. Message

The simple collection request message is sent, but the neighboring node that has sent the topology collection request message to the intermediate node has previously received the topology collection request message sent by other nodes, so the topology collection failure response is returned to the intermediate node. Message.

S203b: After receiving all the topology collection response messages (including the topology collection failure response message and the topology collection success response message), the intermediate node combines the topology information included in the response message with the topology information of the local node, Generating network topology information of the responsible portion of the intermediate node;

It is assumed that the intermediate node sends a topology collection request message to the N neighboring nodes, and receives the topology collection failure response message returned by the M nodes, then after receiving the topology collection success response message returned by the NM nodes, the middle The node learns that all topology collection response messages have been received.

S204b) The intermediate node includes the network topology information of the responsible part in the topology collection response message and returns it to the previous node to complete the topology information collection task of the node.

The main processing steps of the stub node include:

S201c) after receiving the topology collection request message sent by the previous node (which may be an intermediate node or a starting node), in order to avoid a broadcast storm, the stub node verifies whether the message is the topology collection request message received for the first time; If the topology collection request message is received for the first time, the peripheral node returns a topology collection failure response message to the previous node, and ends the processing process of the terminal node; otherwise, the next step is performed;

S202c) The stub node queries the routing table of the node, and sends a topology collection request message to the neighboring node other than the node that has sent the topology collection request message to the stub node, but all the topology collection request is received. If the neighboring nodes of the message return a topology collection failure response message, the node determines that it is a stub node; S203c) The stub node returns a topology collection response message to the node. The network topology information in the message may include only the identifier used to indicate that the node is a stub node, and complete the topology information collection task of the node.

The above identifier may be the node identifier itself of the stub node.

In addition, the stub node can also return all the topology information contained in the routing table of the node to the previous node.

In addition, if a node has no other neighboring nodes except the node to which the topology collection request message is sent, the node is also a stub node.

3 is a schematic diagram showing a topology of a network for collecting topology information by using the method according to the first embodiment of the present invention; as shown in FIG. 3, the arrows of the rows sequentially indicate the distribution level of the topology collection request message, which are respectively the first The level is distributed to the fourth level distribution. The solid arrows in Fig. 3 indicate the first distribution of the topology collection request message; the dotted arrow indicates that the topology collection request message is again transmitted to the node that has received the topology collection request message.

In Fig. 3, node 1 is a start node, and a topology collection request message is sent to node 2, node 3, node 4, and node 5. The node 4 is an intermediate node, and the topology collection request message is sent to the node 6, the node 9, and the node A. The node 6 has been slave node 5 before receiving the topology collection request message sent by the node 4. Upon receiving the topology collection request message having the same topology information collection identifier, the node 6 returns a topology collection failure response message to the node 4. When the node 9 receives the topology collection request message and prepares to send it to the node 8 and the node B, the node 8 and the node B have respectively obtained the topology collection request message from the node 6 and the node A, so the node 9 is returned to the node 9 Pu collects the failure response message, so node 9 becomes a member of the stub node. The specific flow in the scenario shown in Figure 3 will be explained in detail below.

FIG. 4 is a message interaction diagram of the topology information collected by the originating node in the scenario shown in FIG. 3 according to the first embodiment of the present invention, which includes the following steps:

S401) ~ S404) The starting node (node 1) checks the routing table of the node and finds 4 The adjacent nodes are node 2, node 3, node 4, and node 5, respectively, and then send topology collection request messages with the same topology information collection identifier to the four adjacent nodes, respectively, and node 1 stores one record, record The request message is initiated by the node 1 (that is, the local node is recorded as the starting node); the above information recorded by the starting node can be deleted after a preset time (for example, 3600 seconds), and the corresponding information recorded by the following intermediate node and the peripheral node It can also be deleted after a preset time.

S405) After successfully sending the topology collection request message, the starting node enters a topology collection response waiting state, and waits for the topology collection response message returned by the four neighboring nodes;

S406) - S409) after completing the topology collection, the four neighboring nodes respectively return to the node 1 a topology collection response message carrying part of the network topology information;

S410) The node 1 combines the topology information included in the received topology collection response message with the topology information in the routing table of the local node into a complete overlay network topology information, and completes the collection process of the entire network topology information. .

FIG. 5 is a message interaction diagram of topology information collection by an intermediate node in the scenario shown in FIG. 3 according to the first embodiment of the present invention, including the following steps:

S501) the intermediate node (node 4) receives the topology collection request message sent by the previous node (node 1);

S502) The node 4 extracts the topology information collection identifier from the topology collection request message, and searches for the same identifier locally by using the identifier as an index. In this step, the node 4 fails to find the same identifier, and the node 4 determines the topology. The collection request message is received for the first time, and the node 4 stores the identifier, and the record has received the topology collection request message;

S503) ~ S505) Node 4 checks the routing table of the node and finds that there are three neighboring nodes besides node 1, which are node 6, node 9, and node A, respectively, and then send the topology to the three adjacent nodes respectively. The request message is collected, and the topology information collection identifier in the message is consistent with that sent by the node 1; S506) after the topology collection request message is successfully sent, the node 4 enters the topology collection response waiting state, and waits for the three adjacent nodes to return to the extension. Park collects response messages;

S507) ~ S509) after the top three neighbor nodes complete the topology information collection, return to the node 4 a topology collection response message carrying part of the network topology information; S510) The node 4 combines the topology information included in the received topology collection response message with the topology information in the local node routing table into partial topology information;

S511) The node 4 includes the partial topology information generated by the combination and returns it to the previous node (node 1) in the topology collection response message, and the node 4 completes the topology collection task.

FIG. 6 is a message interaction diagram of topology information collection by a stub node in the scenario shown in FIG. 3 according to the first embodiment of the present invention, including the following steps:

S601) The node 9 receives the topology collection request message sent by the previous node (node 4);

S602: The node 9 extracts the topology information collection identifier from the topology collection request message, and searches for the same identifier locally by using the identifier as an index. Here, the node 9 fails to find the same identifier, and determines that the topology collection request message is the first one. Receiving, the node 9 stores the identifier, and the record has received the topology collection request message;

S603) ~ S604) The node 9 checks the routing table of the node, and finds that there are two adjacent nodes in addition to the node 4, namely node 8 and node B, respectively, and then send a topology collection request message to the two adjacent nodes respectively. The topology information collection identifier in the message is consistent with that sent by node 4.

S605) After successfully sending the topology collection request message, the node 9 enters the topology collection response waiting state, and waits for the two neighboring nodes to return the topology collection response message;

S606) Similar to step S602, after receiving the topology collection request message, the node 8 detects the topology information collection identifier in the message, and finds that the topology collection request message with the same topology information collection identifier has been received;

S607) The node 8 returns a topology collection failure response message to the node 9, informing the node 9 that the node has received the topology collection request message, and the node 8 does not further process the message;

After receiving the topology collection request message, the node B detects the topology information collection identifier in the message, and finds that the topology collection request message with the same topology information collection identifier has been received.

5609) Node B sends a topology collection failure response message to node 9, informing node 9 that the node has received the topology collection request message, and node B does not further process the message;

S609) After receiving the topology collection failure response message returned by the two neighboring nodes, the node 9 finds All neighboring nodes have received and processed the topology collection request message, so it is determined that the node is a stub node;

S610) The node 9 returns a topology collection response message to the node 4. The network topology information in the message may only include an identifier for indicating that the node is a stub node, and the node 9 completes the topology collection task.

Of course, the node 9 can also return all the topology information contained in the routing table of the node to the node.

4.

Second embodiment

When there are a lot of nodes in the overlay network, many hops may be needed from the start node to the stub node (that is, many intermediate nodes are included between the start node and the stub node), and the more hops, the start node. It will take a long time to receive the topology collection response message; at the same time, the more nodes included in the overlay network mean that the information amount of the topology information is larger, and the method described in the first embodiment of the present invention is used to collect the extension. In the case of the simple information, the topology collects the response message, so that the message is collected. Therefore, for the overlay network including many nodes, the method also proposes to limit the number of hops collected by a single topology, and collect the entire network through multiple topology collection. Optimization of topology information.

FIG. 7 is a flowchart of a method for collecting topology information of a distributed network according to a second embodiment of the present invention, the method comprising the following steps:

S701) The starting node (node 1) checks the routing table of the node, and finds that there are four neighboring nodes, namely node 2, node 3, node 4, and node 5, and then respectively send the same to the four neighboring nodes. The topology collection request message of the topology information collection identifier (here, for convenience of description, only node 4 is shown in FIG. 7); at the same time, node 1 identifies the maximum number of times of transmission of the message in the topology collection request message (ie, allowed hopping) Number) is N times;

S702: After receiving the topology collection request message, and determining that the message is not the repeated received topology collection request message, the node 4 forwards the message to the next node, and keeps the topology information collection identifier unchanged, and the message is included in the message. The maximum number of passes minus 1 becomes N-1 times;

S703) when the topology collection request message passes N-1 times, when the maximum number of times of transmission in the message becomes 1, the message is sent to the node H; S704) After the node H determines that the message is not the repeatedly received topology collection request message, it is found that the neighboring node (node J) needs to forward the topology collection request message to the routing table, but the forwarding is not allowed at this time. Node H can only use itself as a stub node;

S705) - S707) The node H returns the topology information contained in the routing table (in which the node J is the information of the node to which the topology collection request message has not been forwarded) to the previous node through the topology collection response message, the node H The topology information (in which the node J is the information of the node to which the topology collection request message has not been forwarded) is forwarded by multiple nodes, and finally reaches the initiating node;

S708) When the node 1 combines the topology information included in the received topology collection response message with the topology information in the local routing table, it is found that the topology information is missing, and the missing node is the node J;

5709) The originating node sends a topology collection request message having the same topology information collection identifier as that in step S701 to the node J, and identifies the maximum number of times of transmission in the message as N times;

After receiving the topology collection request message sent by the originating node, the node J completes the topology collection task as an intermediate node or a stub node, and returns a topology collection response message to the node 1, and the node 1 collects the topology according to the node J. The response message completes the assembly of the complete overlay network topology information, and completes the collection process of the entire network topology information.

It can be seen from the above that, by using the method provided by the present invention, the starting node can acquire the topology information of the entire overlay network and form a complete overlay network topology diagram, without causing message blocking, message storm, etc. to the overlay network. .

According to the basic principle of the present invention, the above embodiment may also have various transformation manners, for example:

(1) In the foregoing embodiment, after receiving the repeated topology collection request message, each node returns a topology collection failure response message to the sending node of the message; the node that sends the topology collection request message is received according to the received The topology collects the total number of successful response messages and topology collection failure responses to determine whether all topology collection response messages have been received, and after determining that all topology collection response messages have been received, receiving from the first one The sending node to the topology collection request message returns a topology collection response message; In other embodiments of the present invention, after receiving the repeated topology collection request message, the receiving node of the message may directly discard the message without returning a topology collection failure response message to the sending node of the message; In this case, the sending node of the message needs to set a timer. After the timer expires, the combination of the topology information is performed, and the topology collection response message is returned, and the remaining topology collection response message is no longer waiting. When using this method, as long as the timer is set properly, the topology information of the node is usually not missed. Of course, the network topology information in the topology collection response message received after the timer expires may be sent to the upper node by sending the topology collection response multiple times.

If the remaining topology collection response message is not waited after the timer expires, the node corresponding to the topology collection response message that is not received may also be marked as a missing node, so that the initiating node is in accordance with the second embodiment of the present invention. The method then sends a topology collection request message to the missing node.

(2) In the above embodiment, the intermediate node and the stub node return the collected network topology information (ie, the topology collection success response message) to the node to which the topology collection request message is first sent; in the other aspect of the present invention In an embodiment, the intermediate node and the stub node may return the collected network topology information (ie, the topology collection success response message) to any node to which the topology collection request message is sent, for example, the last one sends the extension information thereto. The node that collects the request message.

(3) In the second embodiment of the present invention, the initial value of the maximum number of forwarding times N of the topology collection request message is set by the originating node, and is included in the topology collection request message and sent to the neighboring node to receive The node to the topology collection request message decrements the N value by ((>>> before forwarding the message, and when the value of N is less than or equal to 1, it indicates that the topology collection request message cannot be forwarded;

In other embodiments of the present invention, other identifiers may be set in the topology collection request message to indicate whether to allow forwarding of the topology collection request message; for example, the starting node sets the forwarding count identifier L, and the initial value is 0. The starting node sends the L value to the neighboring node in the topology collection request message, and the node that receives the topology collection request message adds the L value to / ( />0) before forwarding the message. If the value is greater than or equal to the preset maximum number of forwarding times, the topology collection request message cannot be forwarded.

The above maximum forwarding number and forwarding count identifier may be collectively referred to as a forwarding restriction identifier.

In addition, the above forwarding restriction identifier may also be set by any intermediate node in the peer-to-peer overlay network. Industrial Applicability The present invention provides a topology information collection method for a distributed network and a corresponding distributed network. Any node in the distributed network can obtain the entire network topology information by using the method provided by the present invention, and the method is applicable to a distributed network using any topology structure and arbitrary routing algorithm. At the same time, because the topology information collection request of the present invention is hierarchically copied and distributed, and the mechanism of response grading and reporting is reported, the local traffic of the network can be effectively controlled without causing a big impact on the network.

Claims

Claim

A method for collecting topology information of a distributed network, comprising:

Where M > 1.

2. The method of claim 1 further comprising:

If the node X has no other neighboring nodes except the node to which the topology collection request message is sent, the node X includes the network topology information of the node in the topology collection response message, and returns to send the topology collection to the node. One of the nodes requesting the message.

3. The method of claim 1 further comprising:

If the topology information is not included in all the topology collection response messages received by the node X, the node X includes the network topology information of the node in the topology collection response message, and returns a message to the topology collection request message. One of the nodes.

4. The method according to any one of claims 1 to 3, wherein

The node X returns the topology collection response message containing the network topology information to the transmitting node from which the node receives the first topology collection request message.

5. The method of claim 4, further comprising:

After receiving the topology collection request message, the node X determines whether the message is the first topology collection request message received, and if not, returns a topology collection that does not include the network topology information to the sending node of the message. Response message.

6. The method of claim 2 or 3, wherein

If the node X has no other neighboring nodes except the node to which the topology collection request message is sent, or the topology topology information is not included in all the topology collection response messages received by the node X, the node X returns The network topology information only contains the node identifier of node X.

7. The method of claim 1 further comprising:

After receiving the topology collection request message, the node X determines, according to the forwarding restriction identifier included in the message, whether to allow the topology collection request message to be sent to the M neighboring nodes of the node; if not, the node X And sending, by the node that sends the topology collection request message, a topology collection response message, where the topology collection response message includes network topology information, where the network topology information includes that the M neighboring nodes are not yet forwarded to the network topology information. Topologically collecting information of nodes of the request message;

8. The method of claim 7 further comprising:

Before sending the topology collection request message to its neighboring node, the initial node sets the initial value of the forwarding restriction identifier N to NO, and includes the forwarding restriction identifier in the topology collection request message and sends the Its adjacent nodes;

Where w<0 JL N0< Nl , or w>0 and N0> N2.

9. The method of claim 1 wherein

The topology collection request message includes a topology information collection identifier for uniquely identifying a current topology information collection operation.

10. The method of claim 9, wherein

The topology information collection identifier includes: an identifier of the starting node and a message sequence number.

A distributed network, which employs the method of any of claims 1-10.