CN102065112B - Peer-to-peer (P2P) network system and method and related device for establishing the same - Google Patents

Abstract

The invention discloses a peer-to-peer network system, a method for establishing the peer-to-peer network system and related devices, which are used to solve the problem of low scalability of the existing peer-to-peer network. The system includes: an active super node, including a unit for obtaining the routing information of the second common node when it is determined that there is a second common node in the cluster where it is located and storing corresponding resources; When there is an ordinary node that stores corresponding resources, send the search request to other active super nodes, and use the routing information of the third ordinary node fed back by other active super nodes as the unit of the obtained routing information; and the obtained The routing information is sent to the first common node that sends the search request or other main super node units; the common node is used to send the search request to the main super node in the cluster, and send the search request from the main hyper terminal in the cluster Obtain the resource to be searched from the common node corresponding to the route information of

Description

Peer-to-peer network system, method for establishing a peer-to-peer network system, and related devices

technical field

The present invention relates to the field of network technology, in particular to a peer-to-peer network system, a method for establishing a peer-to-peer network system, a common node device in a peer-to-peer network system, a super node device, a device for establishing a peer-to-peer network system and A node device.

Background technique

With the rapid development of network technology and multimedia technology, people put forward higher and higher requirements for network data transmission speed and network performance. Due to the sharp increase in the number of network users, the network application services constructed in the traditional client/server (C/S, Client/Server) manner have been difficult to meet people's requirements.

In order to solve the above problems, peer-to-peer (P2P, Peer-to-Peer) network technology came into being. Peer-to-peer network technology, also known as peer-to-peer network technology, is a technology used between different terminal nodes (or clients) to directly exchange data without going through a relay device. In a peer-to-peer network, terminal nodes can directly connect to other terminal nodes without connecting to a server for browsing and downloading. Peer-to-peer network technology is widely used in streaming media transmission and other fields because of its load balancing and self-adaptive characteristics.

Common resource search methods in peer-to-peer networks mainly include: peer-to-peer network search based on the central directory server, flooding (flooding) search and distributed hash table search.

Please refer to Figure 1, which is a schematic diagram of an existing resource search method based on a central directory server. Each terminal node submits a catalog of locally stored resources (including documents, streaming media videos, pictures, etc.) Address, etc.) and the corresponding relationship between the resources stored in the terminal node. When a terminal node has a resource demand, it initiates a search request to the central directory server, and the central directory server obtains the routing information of other terminal nodes that store the resources requested by the terminal node through the resource index stored by itself, and sends the obtained routing information of other terminal nodes Routing information is fed back to the terminal node that initiated the search request. Thereafter, resource downloads or transfers can occur directly between the end node that initiated the search request and other end nodes where the resource is stored, without the involvement of a central directory server.

Please refer to Figure 2, which is a schematic diagram of flood search. In an unstructured peer-to-peer network, since there is no central directory server, the basic method of resource lookup is blind search similar to flooding. After each terminal node in Figure 2 receives the search request sent by other terminal nodes, when it judges that it does not store the resources requested by the search request, it will send the above-mentioned received search request to all its neighbors The terminal node is further sent by the neighbor terminal node to more indirect neighbor terminal nodes until the terminal node storing the resources requested by the search request is found, or the search fails after reaching the maximum number of search hops allowed by the system. After finding the terminal node that stores the resource requested by the search request, the terminal node that originally sent the search request can directly transfer from the found terminal node that stores the requested resource based on the routing information fed back by the terminal node that stores the requested resource. Download resources. For example, terminal node 1 is the terminal node that sends the search request. After receiving the search request sent by terminal node 1, terminal node 4 determines that it does not store the resources requested by terminal node 1, and then sends the search request to its neighbors The terminal nodes 5 and 6 happen to have the resources requested by the terminal node 1 stored in the terminal node 5, and the terminal node 1 can directly download the required resources from the terminal node 5.

Please refer to Figure 3, which is a schematic diagram of a structured and fully distributed peer-to-peer network. In such a peer-to-peer network, a distributed hash table (DHT, Distributed Hash Table) is used to determine the keyword of the resource and to search for the resource . The address of the terminal node storing the required resource can be purposefully determined according to the keyword of the resource required by the user. In addition to storing storage resources, each terminal node also stores at least two identities of uplink/downlink neighbor terminal nodes and routing information corresponding to the uplink/downlink neighbor terminal nodes.

When storing resources, first perform a hash operation on the identification of the file to be stored (such as the file name) according to a predetermined hash function to obtain the key value corresponding to the file to be stored, and then store the file to be stored in the node ID and the file to be stored. The key value corresponding to the storage file is closest to the terminal node.

When one of the terminal nodes initiates a search request due to the need to search for resources, first perform a hash operation on the identifier of the file to be searched according to a predetermined hash function to obtain the key value corresponding to the file to be searched, and then obtain the corresponding key value from the neighbor terminal node Select one of the neighbor terminal nodes whose node identification is closest to the key value corresponding to the file to be searched, and send the search request to the selected neighbor terminal node.

For example, please refer to FIG. 3 , which is a schematic diagram of a completely distributed network. It contains 5 terminal nodes, namely: terminal node 1 with node ID 000500, terminal node 2 with node ID 000800, terminal node 3 with node ID 000200, terminal node 4 with node ID 000120 and node ID 000060 The terminal node 5. When terminal node 1 needs to search for a file with a key value of 000080, it first compares the identifiers of its neighbor terminal nodes, terminal node 2 and terminal node 3, and the comparison result is the key value corresponding to the identifier of terminal node 3 and the file to be searched If it is closer, the search request carrying the key value corresponding to the file to be searched is sent to the terminal node 3. After the terminal node 3 receives the search request, after similar comparison and forwarding processing, it finally reaches the terminal node 5, and the terminal node The identity of node 5 is the closest to the key value corresponding to the file to be searched, indicating that the file to be searched is stored in terminal node 5, and finally terminal node 5 feeds back its own routing information to terminal node 1, and terminal node 1 can get from The resource to be searched is obtained at the terminal node 5 .

In the first search method, the central directory server needs to undertake all the work of indexing the directory and determining the address of the terminal node that stores the corresponding resource according to the search request of the terminal node. When there are too many terminal nodes in the peer-to-peer network, the central directory server Processing power can become a bottleneck for searches. And, once the central directory server fails, the entire peer-to-peer network will be paralyzed.

Although flood search solves the above-mentioned problems of peer-to-peer network search based on the central directory server, due to the blindness of terminal nodes forwarding search requests to all their neighbor terminal nodes, it will greatly increase the messages in the network. When the scale of the network is too large and the number of terminal nodes in the network is too large, the number of forwarded search requests will increase sharply, seriously occupying the network bandwidth. Therefore, when this search method is used, the scalability of the peer-to-peer network is not good.

The DHT-based search method is unstable in the network situation. For example, when terminal nodes frequently join/exit the network, the relevant terminal nodes need to update their stored routing information, so there is a problem of high maintenance costs, so it is not suitable for the network Large scale, frequent end nodes joining/or exiting the network.

Contents of the invention

An embodiment of the present invention provides a peer-to-peer network system to solve the problem of low scalability of the peer-to-peer network in the prior art.

Correspondingly, the embodiment of the present invention also provides a method for establishing the peer-to-peer network system, a master super node device, a common node device, and a device for establishing the peer-to-peer network system.

The technical scheme that the embodiment of the present invention provides is as follows:

A peer-to-peer network system, including at least two clusters, each cluster contains a master super node and at least one ordinary node, wherein:

The active super node includes a search request that carries the identifier corresponding to the resource to be searched sent by the first ordinary node in the cluster where it is located or other active super nodes, and determines that there is a second ordinary node in the cluster where it is located When the resource corresponding to the identification is stored, a unit for obtaining the routing information of the second common node; Send the request to an active super node other than the active super node that sent the search request, and receive feedback from other active super nodes other than the active super node that sent the search request. The routing information of the third common node corresponding to the resource is used as a unit of the obtained routing information; and the unit of sending the obtained routing information to the first normal node or other active super nodes sending the search request;

The ordinary node is used to send a search request carrying the identifier corresponding to the resource to be searched to the active super node in the cluster where it is located, and obtain the required resources from the ordinary node corresponding to the routing information sent by the active super node in the cluster where it is located. Search resources.

A method for establishing the peer-to-peer network system, comprising:

The first node requesting to join the peer-to-peer network is used as the first active supernode, and the identifier of the node is added to the supernode identifier set;

When other new nodes subsequently request to join the peer-to-peer network, execute:

Select an active super node identification from the super node identification set, and the distance between the super node corresponding to the selected active super node identification and the new node is the distance between each super node corresponding to all identifications in the super node identification set the closest in distance to the new node;

The distance between the super node corresponding to the selected active super node ID and the new node is less than the first threshold, and the number of ordinary nodes in the cluster where the selected active super node ID corresponds to the super node is not greater than the second threshold , add the new node as a normal node to the cluster where the super node corresponding to the selected primary super node ID is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

A primary super node device, including:

The first receiving unit is configured to receive a search request containing an identification corresponding to the resource to be searched from the first ordinary node in the cluster where the active super node device is located or other active super nodes;

A judging unit, configured to judge whether there is a resource corresponding to the identifier included in the search request received by the first receiving unit stored in a second common node in the cluster where the master supernode device is located;

The first obtaining unit is configured to obtain the routing information of the second normal node when the judging unit judges that there is a second normal node in the cluster where the active super node device is located and stores the resource corresponding to the identifier;

The second acquisition unit is configured to send the search request to a host other than sending the search request when the judging unit judges that there is no common node storing the resource corresponding to the identifier in the cluster where the master supernode device is located The active super node other than the active super node, and receive the feedback of the third common node that stores the resource corresponding to the identification from other active super nodes except the active super node that sent the search request routing information, as the obtained routing information;

The sending unit is configured to send the routing information obtained by the first obtaining unit or the second obtaining unit to the first common node or other active super nodes that send the search request.

A common node device, including:

A storage unit for storing the routing information of the active supernode in the same cluster;

The sending unit is configured to send a search request to the super node device in the same cluster according to the routing information stored in the storage unit, the search request includes the identifier of the resource to be searched;

The receiving unit is configured to receive the routing information of other common node devices that store the resources to be searched and corresponding to the search request sent by the sending unit from the super node device;

A downloading unit, configured to download resources to be searched from other common nodes according to the routing information received by the receiving unit;

The second reporting unit is configured to regularly report the parameter value of the predetermined performance parameter of the common node device to the super node in the same cluster.

A device for establishing a peer-to-peer network system, comprising:

The first determination unit is configured to use the first node requesting to join the peer-to-peer network as the first active supernode, and add the node's identifier to the supernode identifier set;

The selection unit is configured to execute when other new nodes subsequently join the peer-to-peer network:

Select an active super node identification from the super node identification set, and the distance between the super node corresponding to the selected active super node identification and the new node is the distance between each super node corresponding to all identifications in the super node identification set the closest in distance to this new node;

The second determination unit is configured to be in the cluster where the distance between the supernode corresponding to the master supernode identifier selected by the selection unit and the new node is less than the first threshold and where the supernode corresponding to the selected master supernode identifier is located When the number of ordinary nodes is not greater than the second threshold, the new node is added as an ordinary node to the cluster where the supernode corresponding to the selected master supernode identifier is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

A node device, comprising:

a judging unit, configured to judge whether the node device is the first node requesting to join the peer-to-peer network system;

The first determining unit is configured to use itself as the first active super node and add its own identity to the super node when the judging unit determines that the node device is the first node requesting to join the peer-to-peer network system set of identities;

The selection unit is configured to select a master super node ID from the super node ID set when the judging unit determines that the node device is not the first node requesting to join the peer-to-peer network system, and the selection The distance between the super node corresponding to the primary super node ID and the node device is the shortest distance between each super node corresponding to all the IDs in the super node ID set and the node device;

The second determination unit is configured to be in the cluster where the distance between the supernode corresponding to the master supernode identifier selected by the selection unit and the new node is less than the first threshold and where the supernode corresponding to the selected master supernode identifier is located When the number of ordinary nodes is not greater than the second threshold, add itself as an ordinary node to the cluster where the supernode corresponding to the selected active supernode identifier is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

The embodiment of the present invention provides a peer-to-peer network with a novel structure, the peer-to-peer network includes at least two clusters and at least one registration server node, wherein each cluster contains a master super node and at least one common node ; The registration server node is used to store a set of super nodes, and the identification and corresponding routing information of each active super node in the set of super nodes, and the set of super nodes includes the active super nodes in each cluster. Based on the peer-to-peer network with the above structure, when searching for resources, the search method is similar to that based on the central directory server in the cluster, and the search method based on the DHT routing table is used between the clusters, which avoids flooding searches and greatly increases the number of resources in the network. There are bottleneck problems in resource search based on message and central directory server, and the problem of poor scalability of peer-to-peer network caused by complicated maintenance of search method based on DHT routing table.

Description of drawings

FIG. 1 is a schematic diagram of an existing resource search method based on a central directory server;

FIG. 2 is a schematic diagram of an existing flood search;

FIG. 3 is a schematic diagram of resource search based on an existing DHT routing table;

FIG. 4 is a schematic structural diagram of a peer-to-peer network proposed by an embodiment of the present invention;

Fig. 5 is based on the peer-to-peer network of structure shown in accompanying drawing 4, carries out the flowchart of resource search;

Fig. 6 is the schematic diagram that increases standby super node in each cluster in the peer-to-peer network of structure shown in accompanying drawing 4;

FIG. 7 is a schematic structural diagram of an active super node device provided by an embodiment of the present invention;

FIG. 8 is a schematic structural diagram of a second acquisition unit in an active super node device provided by an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of an improved primary super node device provided by an embodiment of the present invention;

FIG. 10 is a schematic structural diagram of a common node device provided by an embodiment of the present invention;

FIG. 11 is a schematic structural diagram of an apparatus for establishing a peer-to-peer network system provided by an embodiment of the present invention;

FIG. 12 is a schematic structural diagram of a node device provided by an embodiment of the present invention.

Detailed ways

Since the existing peer-to-peer network has a problem of low scalability, the embodiment of the present invention proposes a peer-to-peer network with a new structure. The peer-to-peer network contains at least one registration server node and at least two clusters, each of which contains a master super node and several ordinary nodes, and the super nodes in each cluster are the same as the existing central directory server In addition to storing the identifiers, routing information and identifiers of stored resources of each common node in the cluster, the identifiers and routing information of super nodes in adjacent clusters are also stored.

When searching for resources, the peer-to-peer network with the above structure adopts a search method based on a search method based on a central directory server and a search method based on a DHT routing table, specifically:

When searching for resources, after receiving the search request sent by the ordinary nodes in the cluster, the super node first judges whether there are other ordinary nodes in the cluster that store the requested resources. When there is no ordinary node in the cluster that stores the requested resources , forward the search request to super nodes in other clusters until there are common nodes in other clusters that store the requested resources.

The main realization principles, specific implementation modes and corresponding beneficial effects that can be achieved of the technical solutions of the embodiments of the present invention will be described in detail below in conjunction with each accompanying drawing.

Please refer to FIG. 4 , which is a schematic structural diagram of a peer-to-peer network system proposed by an embodiment of the present invention, wherein circles are common nodes, squares are main super nodes, and a registration server node.

Attached Figure 4 contains 3 clusters, each cluster contains a primary super node and at least one common node peer, and for each cluster, the primary super node in the cluster stores the information of each common node in the cluster identification, corresponding routing information, and identification information of resources stored by the common node, and a distributed hash table DHT routing table, the DHT routing table includes the identification and corresponding routing information of the main super node in other clusters, For example, the identity of the active super node in the adjacent cluster and the corresponding routing information, where the routing information may be the IP address, port and other information of the node. For example, ordinary nodes peerA1, peerA2 and master supernode A form cluster A; common nodes peerB1, peerB2, peerB3 and master supernode B form cluster B; peerC1, peerC2, peerC3 and master supernode C form cluster C. In cluster A, the primary supernode A stores the identity of peerA1, the routing information of peerA1, and the identity of resources stored in peerA1, the identity of peerA2, the routing information of peerA2, and the identity of resources stored in peerA2, and peerA1 , peerA2 respectively store the routing information of the master super node in the same cluster - the master super node A. The situation of other clusters is similar and will not be repeated here.

The registration server node super node identification set, the super node set includes the main super node identification in each cluster. It is also possible to store the routing information of the master supernode corresponding to the identifier of each supernode in the supernode identifier set. The registration server is used to provide super node information for new terminal nodes when new terminal nodes join the peer-to-peer network.

The peer-to-peer network system shown in Figure 4 is established in the following manner:

When the first new node requests to join the peer-to-peer network, the new node is used as the first active super node, the identification of the first active super node is determined, and the identification of the first active super node Adding to the super node identification set may preferably store the location information or routing information of the first active super node.

Afterwards, when a new node requests to join the peer-to-peer network, the identification of the new node is determined, and a main super node identification is selected from the set of super node identifications, and the super node identification corresponding to the selected main super node identification The distance between the node and the new node is the closest among the distances between the active super nodes corresponding to all the identities in the super node identification set and the new node.

When the above distance is a geographical distance, the position information of each active super node in the super node identification set can be determined based on the location information of the super node corresponding to each active super node identification in the stored super node identification set and the position information of the new node. The geographical distance between the super node corresponding to the node identifier and the new node, and then select the super node with the closest distance;

When the above distance is a geographical distance, it can be determined based on the routing information of the super node corresponding to each active super node identification in the stored super node identification set and the routing information of the new node to determine the location of each active super node in the super node identification set. The node identifies the network distance between the corresponding super node and the new node, and then selects the closest super node, and the network distance is a function of the round-trip time of the data packet and the hop number of the network node. The new node obtains the identifiers and corresponding routing information of all super nodes in the super node set from the registration server node, and based on the routing information of each super node, conducts a network message round-trip time test to determine the new node and the super node set. The network distance of each super node.

After the new node selects the super node with the closest distance to the node to be new, it judges whether the distance between the selected super node and the terminal node to be allocated is less than the first predetermined threshold D; if so, the new node is added to the selection as a normal node Store the identity of the terminal node to be allocated and the corresponding routing information in the selected super node; otherwise, use the terminal node to be allocated as the active super node in a new cluster, and set The identity of the to-be-new node is added to the super node identity set, and the identity of the terminal node to be allocated and the corresponding routing information are stored. The terminal node to be allocated establishes its own DHT routing table, and the directly adjacent clusters The super node modifies its own DHT routing table.

When the new node is used as the main super node in a new cluster, it can establish its own DHT routing table in the following way: first, based on the routing information of each super node in the super node identification set saved by the registration server node, send to the The super node in the super node identification set sends a message for obtaining the identification of the super node, and according to the obtained identification of each super node, identifies at least two other active super nodes that are closest to its own identification value. The cluster serves as an adjacent cluster, and saves the identifiers and corresponding routing information of the closest at least two other active supernodes;

It is also possible to directly obtain the identifiers of other active supernodes from the registration server node, select at least two clusters of other active supernodes that are closest to their own identifier values as adjacent clusters, and save the closest IDs and corresponding routing information of at least two other active supernodes.

Afterwards, the closest at least two other active supernodes are required to regard themselves as supernodes in their adjacent clusters, that is, the downlink nodes found are required to modify the information of their uplink nodes, and use the new terminal node to be allocated as their uplink node. The node saves the identification and routing information of the terminal node to be allocated; and requires the found upstream node to modify the information of its downstream node, and uses the new node as its downstream node to store the identification and routing information of the terminal node to be allocated.

The above is an example of a new node requesting to join the network to introduce the method of establishing or expanding the peer-to-peer network system. The same process can also be implemented by the registration server, as follows:

The registration server takes the first node requesting to join the peer-to-peer network as the first active supernode, and adds the node's identifier to the supernode identifier set;

When other new nodes subsequently request to join the peer-to-peer network, execute:

Select an active super node identification from the super node identification set, and the distance between the super node corresponding to the selected active super node identification and the new node is the distance between each super node corresponding to all identifications in the super node identification set the closest in distance to this new node;

The distance between the super node corresponding to the selected active super node ID and the new node is less than the first threshold, and the number of ordinary nodes in the cluster where the selected active super node ID corresponds to the super node is not greater than the second threshold , add the new node as a normal node to the cluster where the super node corresponding to the selected primary super node ID is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

When an ordinary node exits the network, the active super node in the cluster where the ordinary node is located shall delete the stored identification, routing information and stored resource identification of the ordinary node.

When an active super node exits the network, the adjacent upstream active super nodes and downlink active super nodes of the active super node should search for other active super nodes with the closest identification to re-establish the routing table; or In a network such as , all active supernodes save the identification and routing information of the above-mentioned directly adjacent uplink/downlink active supernodes, and also save the indirect adjacent uplink/downlink active supernodes within n hops. Identification and routing information, so that when the directly adjacent primary supernode exits the network, you can directly try to connect to the primary supernode that is closest (1 hop) to the primary supernode that exited the network. If the connection is successful , the master super node that is closest (1 hop) to the master super node exiting the network can be used as the updated directly adjacent uplink/downlink master super node, and the routing table can be modified accordingly.

Preferably, the new node requesting to join the network can perform a hash operation on its own routing information (such as IP address, port identifier, etc.) based on a predetermined hash function, and use the result of the hash operation as the identifier of the terminal node. The registrar node may also use a similar manner to determine the identity of the terminal node to be allocated to the newly joined network.

Preferably, in order to ensure that the size of a cluster will not be too large, that is, the number of ordinary nodes in a cluster will not be too large, it is also possible: when the distance between the selected active super node and the new node is less than the first predetermined threshold D. When the number of ordinary nodes in the cluster where the selected master super node is located is not more than the second predetermined threshold G, add the new node to the cluster where the selected master super node is located; When the distance between the super node and the new node is less than the first predetermined threshold D, and the number of ordinary nodes in the cluster where the selected active super node is located is more than the second predetermined threshold G, the new node will be used as a new node in the new cluster. The active super node, and the identification of the new node is added to the super node identification set.

In addition to network distance and geographical distance, other indicators can also be selected, such as message round-trip time, as the basis for new nodes to join the cluster, which will not be listed here.

In the peer-to-peer network system established by the above method, when a new node joins or exits the network, when the terminal node is an ordinary node, only the super node in the cluster where the ordinary node is located needs to modify the stored information; When the node is the active super node, the super nodes in the adjacent clusters are required to modify the stored information and remove the node’s identity from the super node identity set, which is the same as the third fully distributed peer-to-peer in the prior art Compared with the network, the maintenance mechanism is simplified, and when most terminal nodes join or exit the network, it will not cause a large number of terminal nodes to modify the stored information.

Please refer to Figure 5, which is a schematic diagram of the resource search process based on the peer-to-peer network system shown in Figure 4. When performing resource search, the super node acts as an index server (tracker), and the specific search process is as follows:

Step 51, when PeerA1 needs to search for a file whose resource identifier is X, it sends a search request to the active supernode A in the same cluster, and the search request includes the identifier X of the resource to be searched;

Step 52, supernode A judges whether there are other common nodes in the cluster that store For the file marked as X, if so, proceed to step 53, otherwise proceed to step 55;

Step 53, the active supernode A has other ordinary nodes in the cluster, for example, when peerA2 stores the resource searched by peerA1, that is, when the identifier of the stored resource is consistent with the identifier of the resource contained in the search request, it will store the resource identifier The routing information of ordinary node peerA2 of X is sent to peerA1;

Step 54, peerA1 can obtain the resources to be searched from peerA2 according to the routing information of peerA2;

Step 55: When there is no resource identifier stored by a common node in the cluster that is consistent with the resource identifier contained in the search request, the master supernode A sends the search request to other clusters, such as the master supernode in the adjacent cluster. Use a super node, such as the main super node B or the super main level node C, go to step 45;

Step 56, assuming that after the active super node A sends the search request to the active super node B, the active super node B, according to the identification of each common node in the cluster B stored by itself, the routing information and the stored resource identification relationship, the identifier of the resource contained in the received search request, and when it is determined that peerB1 stores the resource that peerA1 wants to search for, send the routing information of peerB1 to the active supernode A;

Step 57, the active super node A sends the received routing information of the common node peerB1 storing the resource to be searched sent by the active super node in the adjacent cluster to peerA1;

In step 58, peerA1 can obtain the resource to be searched from peerB1 according to the routing information of peerB1.

Preferably, the active super node A selects an active super node that forwards the search request from the active super node B and the active super node C in the adjacent cluster in the following manner: the active super node A follows a predetermined Hash algorithm, perform hash operation on the resource identifier X contained in the search request, and select the corresponding super node identifier from the stored DHT routing table in the adjacent cluster that is closest to the hash operation result Y and send the search request to the selected active SuperNode B according to the routing information of the selected active SuperNode B.

Preferably, in the peer-to-peer network system shown in FIG. 4 , ordinary nodes regularly report the identifiers of their currently stored resources to the active supernodes in the cluster, so that the active supernodes in the cluster can update in time The identification of each common node in the cluster stored by itself, the corresponding routing information and the identification information of the resources stored by the common node. For example, peerA1 and peerA2 should periodically report the identifiers of the resources they currently store to the active supernode A.

Preferably, please refer to accompanying drawing 6, in order to improve the security of the peer-to-peer network system shown in accompanying drawing 4, in order to avoid in a cluster, because the primary super node breaks down, or withdraws from the network and cause the common node in the cluster Resource search is not possible. In addition to the main super node, each cluster can contain at least one backup super node. The backup super node stores the same information as the main super node, that is, the route of each common node in the cluster The corresponding relationship between the information and the identifier of the resource stored by the ordinary node, and the distributed hash table DHT routing table, which contains the identifier of the super node in other clusters and the corresponding routing information; and each cluster in the cluster Each common node stores the identification and routing information of at least one standby supernode in the cluster, so that when the active supernode fails, resource search can be performed through the standby supernode. The active supernode and standby supernode in each cluster periodically synchronize information to ensure the consistency of the stored information. Ordinary nodes regularly update the parameter values of their own performance parameters, including online time, free storage space, information processing capabilities, network stability, etc., so that the master supernode can change from A backup super node is selected from ordinary nodes. Before an ordinary node is selected as a backup super node, the user of the ordinary node should be reminded whether to agree, because being a backup super node will occupy part of the storage space and processing power of the ordinary node. In addition to the failure of the active super node, in some other predetermined situations, such as when the processing capacity or storage space of the active super node drops beyond the mobile threshold, the standby super node can also actively work as the active super node.

In the case of a backup super node, in the above-mentioned solution for establishing a peer-to-peer network, the registration server node should store the identification and routing information of the active super node.

The above-mentioned search method adopted by the peer-to-peer network proposed by the embodiment of the present invention is a search method based on the combination of the search method based on the central directory server and the search method based on the DHT routing table, which avoids a large increase in the flood search method in the network. Disadvantages of messages and occupying a large amount of network bandwidth. At the same time, only ordinary nodes in each cluster rely on super nodes in the cluster (including active super nodes and standby super nodes) when searching, even if a super node in a cluster fails and goes down , ordinary nodes in other clusters will not be affected, which largely solves the bottleneck and poor security of the existing peer-to-peer network search based on the central directory server, as well as the poor scalability caused by the bottleneck good question.

Taking the P2P network streaming media service as an example, the process of resource search based on the peer-to-peer network system shown in Figure 4 is introduced below:

PeerA1 needs to search for a video file named "News Network 2009-7-1.rmvb". First, peerA1 sends a search request to the main super node A, which contains the name of the resource to be searched "News Network 2009- 7-1. rmvb".

The master supernode A judges the cluster according to the identifier of each common node stored in the cluster, the corresponding relationship between routing information and the identifier of the stored resource (as shown in Table 1), and the identifier of the resource contained in the received search request. Whether other ordinary nodes in the network store a file named "News Network 2009-7-1.rmvb", if peerA2 stores the file, feed back the routing information of peerA2 to peerA1; Hash algorithm, perform hash operation on the name of the resource to be searched "News Network 2009-7-1.rmvb", and get the hash operation result 12. The DHT routing table stored by the main super node A is shown in Table 2.

Table 1 Information of common nodes in the cluster stored by master super node A

common node name routing information ID of the stored resource other information PeerA1 202.111.220.1 News broadcast 2009-7-2.rmvb PeerA2 202.111.220.1 News broadcast 2009-7-3.rmvb

Table 2 Information of the master super node in the adjacent cluster stored by the master super node A

Primary super node name logo routing information other information Primary Super Node B 11 202.111.222.1 Primary Super Node C 20 202.111.222.20

Since the identity of the primary supernode B is the closest to the result of the hash operation, the primary supernode A sends the search request to the selected primary supernode B. According to the identity of each common node in the cluster B stored by the master supernode B, the corresponding relationship between routing information and the identity of the stored resource (as shown in Table 3), and the name of the resource contained in the received search request, When it is determined that peerB1 stores the resource that peerA1 wants to search for, it sends the routing information "202.111.230.1" of peerB1 to the active supernode A.

Table 1. The information of common nodes in the cluster stored by the main SuperNode B

common node name routing information ID of the stored resource other information PeerB1 202.111.230.1 News broadcast 2009-7-1.rmvb PeerB2 202.111.230.2 News broadcast 2009-7-4.rmvb PeerB3 202.111.230.3 News broadcast 2009-7-7.rmvb

The master supernode A sends the received routing information of peerB1 to peerA1, and peerA1 can obtain the resource to be searched from peerB1 according to the routing information of peerB1.

The following takes the VOD (Video On Demand) business as an example to introduce the resource search process:

In addition to video files, resources stored in common nodes can also be slices or chunks in video files.

In the VOD system, when a new node requests to join the peer-to-peer network, the registration server RS first judges whether the video server providing the program source has enough bandwidth to provide services for the new node, and if the video server does not have enough bandwidth for When providing services for a new node, the routing information of the super nodes in the super node set is provided to the new node.

The new node determines itself as an ordinary node in an existing cluster based on the distance from the existing super node, the first predetermined threshold D, and the second predetermined threshold G according to the scheme described in the above-mentioned method of establishing a peer-to-peer network system, or A primary supernode in a new cluster.

In the peer-to-peer network system as shown in Figure 4, the master supernode A, the master supernode B and the master supernode C all store the names of the common nodes in their respective clusters, the corresponding routing information and The identifiers of different video segments chunks stored in the same video file "News Network 2009-7-1.rmvb", the corresponding video segment of the identifier Seg[n] of the video segment is determined by the start time and end time, for example, Seg [n]=[00:01:25,00:03:25] refers to the video data corresponding to the video signal from the first minute 25 seconds to the third minute 25 seconds when the video file is played.

Table 4 shows the information of common nodes in the cluster stored by primary super node A.

Table 4 Information of ordinary nodes in the cluster stored by master super node A

common node name routing information ID of the stored resource other information PeerA1 202.111.220.1 Seg[1], Seg[3] PeerA2 202.111.220.1 Seg[3], Seg[4]

PeerA1 needs to search for the second video clip Seg[2] in "News Network 2009-7-1.rmvb". First, peerA1 sends a search request to the active super node A in the same cluster, and the search request contains the The identifier of the video segment Seg[2].

The master supernode A judges the cluster according to the names of the common nodes in the cluster A stored by itself, the corresponding routing information and the identifiers of different video clips stored in the same video file "News Network 2009-7-1.rmvb". Whether other ordinary nodes in A store the second video segment Seg[2] in the file "News Network 2009-7-1.rmvb", if there are other ordinary nodes in cluster A storing this video segment, it will correspond The routing information of the node is fed back to peerA1. Preferably, if there is at least one other common node storing the video segment Seg[2], the routing information list of all common nodes storing the video segment Seg[2] can be provided to peerA1 , peerA1 selects at least one of the common nodes to obtain the second video segment Seg[2]; otherwise, it forwards the search request to the active super node in the adjacent cluster, the specific process is similar to the previous embodiment, and here Let me repeat.

After peerA1 obtains the second video segment Seg[2] in "News Network 2009-7-1. Supernode A should modify the ID of the video segment stored in the related record of peerA1, and the modified peerA1 related information is shown in Table 5.

Table 5 Information of ordinary nodes in the cluster stored by super node A

common node name routing information ID of the stored resource other information PeerA1 202.111.220.1 Seg[1], Seg[2], Seg[3]

Preferably, in the VOD system, the chunks stored in each common node change quickly, and the active supernode in the cluster can obtain the identifier of the resource stored by the common node from the common node more frequently, so as to update its own stored The information of common nodes.

Preferably, when an ordinary node needs the next video segment, before sending a search request to the active super node in the cluster, it can inquire whether other ordinary nodes that have established a download connection have stored the next video segment needed. Ordinary nodes that have established download connections store the next video segment in need, and then download the next video segment from the ordinary nodes that have established download connections preferentially, without sending search requests to super nodes in the cluster.

Correspondingly, the embodiment of the present invention also provides a primary super node device in a peer-to-peer network. As shown in FIG. Acquiring unit 704 and sending unit 705, wherein:

The first receiving unit 701 is configured to receive a search request containing an identification corresponding to the resource to be searched from the first ordinary node in the cluster where the active super node device is located or other active super nodes;

A judging unit 702, configured to judge whether there is a resource corresponding to the identifier included in the search request received by the first receiving unit 701 stored in a second common node in the cluster where the active supernode device is located;

The first obtaining unit 703 is configured to obtain the routing information of the second normal node when the judging unit 702 judges that there is a second normal node in the cluster where the active super node device is located and stores the resource corresponding to the identifier;

The second obtaining unit 704 is configured to send the search request to the server other than sending the Search for active super nodes other than the active super node that sent the search request, and receive feedback from other active super nodes other than the active super node that sent the search request, and store the resources corresponding to the identification. The routing information of the node, as the obtained routing information;

The sending unit 705 is configured to send the routing information obtained by the first obtaining unit 703 or the second obtaining unit 704 to the first common node or other active super nodes that send the search request.

Specifically, the judging unit 702 compares the stored resource identifiers and search request An ordinary node with the same identifier carried in the ID is used as the second ordinary node storing the resource corresponding to the identifier; and

When there is no identifier of the resource stored by the common node that is consistent with the identifier carried in the search request, it is determined that there is no common node that stores the resource corresponding to the identifier in the cluster where it is located.

Preferably, referring to FIG. 8, the second acquiring unit 704 specifically includes a selection subunit 801, a sending subunit 802, a receiving subunit 803, and a determining subunit 804, wherein:

The selection subunit 801 is configured to select an active super node other than the active super node that sent the search request, the search request being the search request received by the first receiving unit;

a sending subunit 802, configured to send the search request received by the first receiving unit 701 to the active super node selected by the selecting subunit 801;

The receiving subunit 803 is configured to receive the routing information of the third common node that stores the resources corresponding to the identifier and is fed back by the active super node selected by the selecting subunit 801 according to the search request sent by the sending subunit 802;

The determining subunit 804 is configured to use the routing information received by the receiving subunit 803 as the obtained routing information.

Preferably, the selection subunit 801 specifically includes an operation module and a selection module, wherein:

An operation module, configured to perform a hash operation on the identifier of the resource contained in the search request received by the receiving unit according to a predetermined hash algorithm;

The selection module is used to select from the stored DHT routing table the second active supernode whose corresponding identifier is closest to the hash operation result of the operation module, and the DHT routing table contains other active supernodes except itself. The identification of the super node and the corresponding routing information.

Preferably, please refer to accompanying drawing 9, the active super node device in accompanying drawing 7 also includes the second receiving unit 901 and the selecting unit 902, wherein:

The second receiving unit 901 is configured to receive the parameter value of the performance parameter of the common node regularly reported by the common nodes in the same cluster;

The selection unit 902 is configured to select at least one common node as a standby super node according to the parameter value received by the second receiving unit 901 .

Preferably, the active super node device as shown in FIG. 9 further includes an information synchronization unit, configured to regularly perform information synchronization with the backup super node selected by the selection unit.

Please refer to the accompanying drawing 10, which is a schematic structural diagram of a common node device provided by an embodiment of the present invention. The common node device includes a storage unit 110, a sending unit 120, a receiving unit 130, and a downloading unit 140, wherein:

a storage unit 110, configured to store routing information of active super nodes in the same cluster;

The sending unit 120 is configured to send a search request to the super node device in the same cluster according to the routing information stored in the storage unit 110, and the search request includes the identifier of the resource to be searched;

The receiving unit 130 is configured to receive the routing information of other ordinary node devices that store the resources to be searched and corresponding to the search request sent by the sending unit 120 sent by the super node device;

The downloading unit 140 is configured to download the resources to be searched from the other common nodes according to the routing information received by the receiving unit 130 .

Preferably, the common node device in FIG. 10 further includes a first reporting unit, configured to periodically report the identifier of the resource stored by the common node device to the super nodes in the same cluster.

Preferably, the common node device in FIG. 10 further includes a second reporting unit, configured to periodically report the parameter value of the predetermined performance parameter of the common node device to the super nodes in the same cluster.

Please refer to the accompanying drawing 11, which is a device for establishing a peer-to-peer network system provided by an embodiment of the present invention, the device includes a first determination unit 111, a selection unit 112 and a second determination unit 113, wherein

The first determining unit 111 is configured to use the first node requesting to join the peer-to-peer network as the first active supernode, and add the node's identifier to the supernode identifier set;

The selection unit 112 is configured to execute when other new nodes subsequently join the peer-to-peer network:

Select an active super node identification from the super node identification set, and the distance between the super node corresponding to the selected active super node identification and the new node is the distance between each super node corresponding to all identifications in the super node identification set the closest in distance to this new node;

The second determination unit 113 is configured to select the location where the distance between the supernode corresponding to the master supernode identifier selected by the selection unit 112 and the new node is less than the first threshold, and where the supernode corresponding to the selected master supernode identifier is located. When the number of ordinary nodes in the cluster is not greater than the second threshold, the new node is added as an ordinary node to the cluster where the supernode corresponding to the selected master supernode identifier is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

The apparatus for establishing a peer-to-peer network system shown in FIG. 11 may be the registration server in the method embodiment.

Please refer to Figure 12, which is a schematic structural diagram of a node device provided by an embodiment of the present invention. The node device includes a judgment unit 121, a first determination unit 122, a selection unit 123, and a second determination unit 124, wherein:

A judging unit 121, configured to judge whether the node device is the first node requesting to join the peer-to-peer network system;

The first determining unit 122 is configured to, when the judging unit 121 judges that the node device is the first node requesting to join the peer-to-peer network system, use itself as the first active super node, and add its own identity to the peer-to-peer network system. super node identification set;

The selection unit 123 is configured to select a master supernode identifier from the supernode identifier set when the judging unit 121 judges that the node device is not the first node requesting to join the peer-to-peer network system. The distance between the supernode corresponding to the main supernode identifier selected above and the node device is the closest among the distances between each supernode corresponding to all the identifiers in the supernode identifier set and the node device;

The second determination unit 124 is configured to select the location where the distance between the supernode corresponding to the active supernode identifier selected by the selection unit 123 and the new node is less than the first threshold, and where the selected active supernode identifier corresponds to the supernode. When the number of ordinary nodes in the cluster is not greater than the second threshold, add itself as an ordinary node to the cluster where the supernode corresponding to the selected master supernode identifier is located; otherwise

The new node is used as the active super node, and the identification of the new node is added to the super node identification set.

Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims (23)

1. a peer-to-peer network system is characterized in that, comprises at least two bunches, comprises a primary super node and at least one ordinary node in each bunch, wherein:

Primary super node, comprise for the searching request of wanting sign corresponding to searching resource of carrying of sending according to first ordinary node at self place bunch or other primary super nodes, when in determining self place bunch, existing the second ordinary node to store resource corresponding to described sign, obtain the unit of the routing iinformation of described the second ordinary node; And when in determining self place bunch, not having the ordinary node that stores resource corresponding to described sign, described searching request is sent to primary super node except the primary super node that sends described searching request, and receive the routing iinformation of the 3rd ordinary node that stores resource corresponding to described sign of other the primary super nodes feedbacks except the primary super node that sends described searching request, as the unit of the routing iinformation of acquisition; And the routing iinformation that obtains sent to the first ordinary node of sending described searching request or the unit of other primary super nodes;

Ordinary node, be used for primary super node in self place bunch and send the searching request of carrying sign corresponding to the resource that will search for, and obtain the resource that to search for ordinary node corresponding to the routing iinformation that sends of the primary super node in self place bunch.

2. peer-to-peer network as claimed in claim 1 system, it is characterized in that, described primary super node specifically is used for according to the routing iinformation of each ordinary node in self pre-stored place bunch and the corresponding relation of the sign of the resource of this ordinary node storage, and the ordinary node that the sign of carrying in the sign of the resource of storage and the searching request is consistent is as storing described the second ordinary node that identifies corresponding resource; And the sign of carrying in the sign of the resource that does not have ordinary node storage and searching request is when consistent, determines not exist in self place bunch the ordinary node of the resource that stores described sign correspondence.

3. peer-to-peer network as claimed in claim 1 system, it is characterized in that, described primary super node is selected a primary super node except the primary super node that sends described searching request, and described searching request is sent to the primary super node of selecting.

4. peer-to-peer network as claimed in claim 3 system, it is characterized in that, described primary super node specifically is used for according to predetermined hash algorithm, sign to the resource that comprises in the described searching request is carried out Hash operation, and from the distributed hashtable DHT routing table of storage, select corresponding sign and Hash operation result the immediate second primary super node, comprise sign and the corresponding routing iinformation of other the primary super nodes except self in the described DHT routing table.

5. peer-to-peer network as claimed in claim 1 system is characterized in that, also comprises super node for subsequent use in each bunch, be used for termly with bunch primary super node carry out information synchronization, and under predetermined condition, replace primary super node work.

6. peer-to-peer network as claimed in claim 5 system is characterized in that, described ordinary node also be used for termly the sign of the resource of self storage or/and the parameter value of self performance parameter report with bunch primary super node;

Described primary super node also be used for according to bunch the parameter value of the performance parameter that reports of ordinary node, in self place bunch, select at least 1 ordinary node as super node for subsequent use.

7. peer-to-peer network as claimed in claim 6 system is characterized in that, the performance parameter of described ordinary node comprises line duration, idle storage space, information processing capability, the network bandwidth or network stabilization.

8. a method of setting up peer-to-peer network as claimed in claim 1 system is characterized in that, comprising:

First node of asking to add described peer-to-peer network as first primary super node, is gathered the sign adding super node sign of described node;

When follow-up other new node requests add described peer-to-peer network, carry out:

Select a primary super node sign from the set of described super node sign, super node corresponding to the described primary super node sign of selecting and the distance of this new node are that all identify in the distance of each corresponding super node and this new node nearest in the set of super node sign;

The distance of super node corresponding to the primary super node sign of selecting and this new node super node place corresponding less than first threshold and the primary super node sign selected bunch in the quantity of ordinary node when being not more than Second Threshold, the primary super node that this new node is brought Selection In out as ordinary node identify correspondence the super node place bunch in; Otherwise

This new node as primary super node, is gathered the sign adding super node sign of this new node.

9. method as claimed in claim 8 is characterized in that, when follow-up other new nodes add described peer-to-peer network, selects a primary super node sign from the set of described super node sign, specifically comprises:

Based on the routing iinformation of the primary super node of the sign correspondence of each node in the pre-stored super node sign set, determine the network distance of the primary super node that described new node is corresponding with each sign in the set of super node sign; And

Select the sign of a primary super node, super node corresponding to the described primary super node sign of selecting and the network distance of this new node are that all identify in the network distance of each corresponding super node and this new node nearest in the set of super node sign.

10. method as claimed in claim 8 is characterized in that, when follow-up other new nodes add described peer-to-peer network, selects a primary super node sign from the set of described super node sign, specifically comprises:

Based on the primary super positional information of the sign correspondence of each node in the pre-stored super node sign set, determine the geographic distance of the primary super node that described new node is corresponding with each sign in the set of super node sign; And

Select the sign of a primary super node, super node corresponding to the described primary super node sign of selecting and the geographic distance of this new node are that all identify in the geographic distance of each corresponding super node and this new node nearest in the set of super node sign.

11. method as claimed in claim 8 is characterized in that, after the sign adding super node sign set with described new node, also comprises the DHT routing table of storing in definite this new node.

12. method as claimed in claim 11 is characterized in that, determines specifically to comprise the DHT routing table of storing in this new node:

Determine the sign of described new node, and

Obtain the sign of the primary super node that comprises in the set of super node sign, and

In the sign of the primary super node that obtains, choose immediate at least two other primary super nodes on the sign numerical value with described new node, and in this new node, preserve sign and the corresponding routing iinformation of at least two other primary super nodes choosing.

13. method as claimed in claim 11 is characterized in that, after the DHT routing table of storing in determining this new node, also comprises:

Revise the DHT routing table of storing in primary super node corresponding to part sign in the set of super node sign, primary super node corresponding to described part sign is the primary super node of the sign correspondence that comprises in the DHT routing table of storing in the described new node.

14. method as claimed in claim 13 is characterized in that, revises the DHT routing table of storing in primary super node corresponding to described part sign, specifically comprises:

The sign and the corresponding routing iinformation that add this new node in the DHT routing table of in primary super node corresponding to described part sign, storing.

15. such as the described method of arbitrary claim in the claim 8 to 14, it is characterized in that, determine that in the following manner request adds the sign of the new node of described peer-to-peer network:

Obtain the IP address of described new node, and

Hash function based on predetermined carries out Hash operation to the IP address that gets access to, and

With the result of the described Hash operation sign as described new node.

16. a method of setting up the peer-to-peer network system is characterized in that, comprising:

The new node of the described peer-to-peer network of request adding is determined when self being the node of the described peer-to-peer network of first request adding system, and self as first primary super node, is gathered the sign adding super node sign of self; And

Self not first request when adding the node of described peer-to-peer network system determining, carry out:

Described new node is selected a primary super node sign from the set of described super node sign, super node corresponding to the described primary super node sign of selecting and the distance of this new node are that all identify in the distance of each corresponding super node and this new node nearest in the set of super node sign; And

The distance of super node corresponding to the primary super node sign of selecting and this new node super node place corresponding less than first threshold and the primary super node sign selected bunch in the quantity of ordinary node when being not more than Second Threshold, the primary super node that self brings Selection In out as ordinary node is identified correspondence the super node place bunch in; Otherwise

This new node as primary super node, is gathered the sign adding super node sign of this new node.

17. a primary super node equipment is characterized in that, comprising:

The first receiving element is used for receiving the searching request of wanting sign corresponding to searching resource that comprises that first ordinary node at described primary super node equipment place bunch or other primary super nodes send;

Judging unit is used for judging whether described primary super node equipment place bunch exists the second ordinary node to store the resource of the sign correspondence that comprises in the searching request that the first receiving element receives;

The first acquiring unit is used for obtaining the routing iinformation of described the second ordinary node when judgment unit judges goes out described primary super node equipment place and bunch exists second ordinary node to store resource corresponding to described sign;

Second acquisition unit, be used for when judgment unit judges goes out described primary super node equipment place and bunch do not have the ordinary node that stores resource corresponding to described sign, described searching request is sent to primary super node except the primary super node that sends described searching request, and the routing iinformation of the 3rd ordinary node that stores resource corresponding to described sign of other the primary super nodes feedbacks of reception except the primary super node that sends described searching request, as the routing iinformation of acquisition;

Transmitting element is used for the routing iinformation of the first acquiring unit or second acquisition unit acquisition is sent to the first ordinary node or other the primary super nodes that sends described searching request.

18. primary super node equipment as claimed in claim 17 is characterized in that, described second acquisition unit specifically comprises:

The chooser unit is used for selecting a primary super node except the primary super node that sends described searching request, and described searching request is the searching request that the first receiving element receives;

Send subelement, the searching request that is used for the first receiving element is received sends to the primary super node that the chooser unit is selected;

The reception subelement, the routing iinformation of the 3rd ordinary node of the resource that stores described sign correspondence that the searching request that the primary super node that is used for selecting reception chooser unit sends according to the transmission subelement is fed back;

Determine subelement, be used for to receive routing iinformation that subelement receives as the routing iinformation of acquisition.

19. primary super node equipment as claimed in claim 18 is characterized in that, described chooser unit specifically comprises:

Computing module is used for according to predetermined hash algorithm, and the sign of the resource that comprises in the described searching request that receiving element is received is carried out Hash operation;

Select module, be used for selecting corresponding sign and Hash operation result's immediate second primary super node of computing module from the DHT routing table of storage, comprise sign and the corresponding routing iinformation of other the primary super nodes except self in the described DHT routing table.

20. primary super node equipment as claimed in claim 17 is characterized in that, also comprises:

The second receiving element, be used for to receive with bunch the parameter value of performance parameter of the described ordinary node that regularly reports of ordinary node;

Selected cell for the parameter value that receives according to the second receiving element, is selected at least one ordinary node as super node for subsequent use.

21. an ordinary node equipment is characterized in that, comprising:

Memory cell is for the routing iinformation of storing same bunch primary super node;

Transmitting element is used for the routing iinformation according to cell stores, and the super node equipment in same bunch sends searching request, comprises the sign of the resource that will search in this searching request;

Receiving element is used for receiving the routing iinformation of other ordinary node equipment that store the resource that will search for corresponding to searching request that described super node equipment sends sends with transmitting element;

Download unit for the routing iinformation that receives according to receiving element, is downloaded the resource that will search for from described other ordinary nodes;

Second reports the unit, is used for reporting to same bunch super node termly the parameter value of this ordinary node equipment pre-determined characteristics parameter.

22. a device of setting up the peer-to-peer network system is characterized in that, comprising:

The first determining unit, the node that is used for first request is added described peer-to-peer network are as first primary super node, with the sign adding super node sign set of described node;

Selected cell is used for carrying out when follow-up other new nodes add described peer-to-peer network:

Select a primary super node sign from the set of described super node sign, super node corresponding to the described primary super node sign of selecting and the distance of this new node are that all identify in the distance of each corresponding super node and this new node nearest in the set of super node sign;

The second determining unit, when the distance that is used for super node corresponding to the primary super node sign selected at selected cell and this new node is not more than Second Threshold less than the quantity of bunch ordinary node at super node place corresponding to first threshold and the primary super node sign selected, the primary super node that this new node is brought Selection In out as ordinary node identify correspondence the super node place bunch in; Otherwise

This new node as primary super node, is gathered the sign adding super node sign of this new node.

23. a node device is characterized in that, comprising:

Judging unit is used for judging whether described node device is the node that first request adds the peer-to-peer network system;

The first determining unit is used for when judgment unit judges goes out the node of described node device described peer-to-peer network system for first request adds, and self as first primary super node, is added the super node sign with self sign and gather;

Selected cell, being used for going out described node device in judgment unit judges is not that first request is when adding the node of described peer-to-peer network system, select a primary super node sign from the set of described super node sign, super node corresponding to the described primary super node sign of selecting and the distance of this node device are that all identify in the distance of each corresponding super node and this node device nearest in the set of super node sign;

The second determining unit, when the distance that is used for super node corresponding to the primary super node sign selected at selected cell and this new node is not more than Second Threshold less than the quantity of bunch ordinary node at super node place corresponding to first threshold and the primary super node sign selected, the primary super node that self brings Selection In out as ordinary node is identified correspondence the super node place bunch in; Otherwise

This new node as primary super node, is gathered the sign adding super node sign of this new node.

Images