WO2012136094A1 - Procédé et système de maîtrise de surcharge dans un réseau poste à poste - Google Patents

Procédé et système de maîtrise de surcharge dans un réseau poste à poste Download PDF

Info

Publication number
WO2012136094A1
WO2012136094A1 PCT/CN2012/072273 CN2012072273W WO2012136094A1 WO 2012136094 A1 WO2012136094 A1 WO 2012136094A1 CN 2012072273 W CN2012072273 W CN 2012072273W WO 2012136094 A1 WO2012136094 A1 WO 2012136094A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
message
data
receiving
backup
Prior art date
Application number
PCT/CN2012/072273
Other languages
English (en)
Chinese (zh)
Inventor
陶全军
吴建华
郝振武
胡永生
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2012136094A1 publication Critical patent/WO2012136094A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/563—Data redirection of data network streams
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/12—Applying verification of the received information
    • H04L63/123—Applying verification of the received information received data contents, e.g. message integrity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks
    • H04L67/1044—Group management mechanisms 
    • H04L67/1048—Departure or maintenance mechanisms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/40—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection

Definitions

  • the present invention relates to peer-to-peer (P2P, Peer-to-Peer) network technology, and more particularly to a method and system for controlling overload in a P2P network.
  • P2P peer-to-peer
  • Peer-to-Peer Peer-to-Peer
  • P2P network technology has fully demonstrated strong technical advantages in file exchange, distributed computing, and service sharing.
  • the application of P2P network technology is mainly concentrated in the application of the Internet, but has not been applied on a large scale in the traditional telecommunication network.
  • SIP Session Initiation Protocol
  • the network technology is introduced into the telecommunication network, it is required to ensure the application of SIP in the P2P network. It also requires the ability to run a variety of typical businesses.
  • Figure 1 shows a P2P core network that most researchers and research institutions agree with in existing research.
  • the P2P core network includes: a P2P overlay network 101, a P2P node 102, a non-P2P terminal 103, and a P2P terminal 104.
  • the P2P overlay network 101 is a logical network composed of P2P nodes 102 of different roles; the P2P node 102 is a basic component of the P2P overlay network 101, and is capable of providing storage and transmission to other nodes in the P2P overlay network 101.
  • the node of the service; the non-P2P terminal 103 is a terminal device that does not support any P2P protocol and only supports SIP in the P2P overlay network 101; the P2P terminal 104 supports both the P2P protocol and the P2P overlay network 101.
  • the SIP-enabled terminal device is weak in the terminal device, especially the handheld terminal device. Therefore, in actual deployment, the P2P terminal 104 is only used as a P2P client to access the P2P overlay network 101, and does not assume storage.
  • Server functions such as transmission, and, taking into account the needs of telecommunications operations, Both the P2P terminal 104 and the non-P2P terminal 103 are required to access the P2P overlay network 101 through the nearest P2P node 102.
  • the P2P node 102 When the P2P node 102 simultaneously serves as the access node of the non-P2P terminal 103 and the P2P terminal 104, for non-P2P
  • the terminal 103 the P2P node 102 is responsible for converting the SIP adopted by the non-P2P terminal 103 and the protocol used for routing the SIP inside the P2P overlay network 101, and then relaying the message.
  • the P2P node 102 only needs to message. Relay.
  • the P2P nodes 102 implement dynamic real-time backup of data between nodes according to a Distributed Hash Table (DHT) algorithm.
  • DHT Distributed Hash Table
  • FIG. 2 is a schematic flowchart of a conventional network overload control method. As shown in FIG. 2, the process includes the following steps:
  • Step 201 The forward node senses that the backward node is overloaded.
  • the forward node senses that the backward node is overloaded by: the backward node or the overload server notifies the forward node that the backward node is overloaded, or the forward node Inferring that the backward node is overloaded according to the message delay;
  • Step 202 The forward node discards or sends a message to the backward node according to the predetermined ratio according to the overload policy and the backward node overload, so as to reduce the load of the backward node.
  • the node when the node is aware of the backward node overload, it will discard or send the message to the backward node according to the established ratio.
  • the telecommunications that introduces the P2P network technology are controlled.
  • the overload in the network can not reflect the advantages of P2P network technology, but will increase the instability of P2P network.
  • the advantages of P2P network are: exchange quality for quantity, but single node in P2P network
  • the stability is less stable than the single node in the traditional telecommunication network. Therefore, when the network is under heavy load, the nodes in the P2P network are more likely to form resource overload.
  • the traditional overload control method is used, When the load is high, the connection rate of the telecommunication network that introduces the P2P network technology is far lower than that of the traditional telecommunication network, thus greatly reducing the user experience. Summary of the invention
  • the main object of the present invention is to provide a method and system for controlling overload in a P2P network, which can ensure continuity of telecommunication services and thereby improve user experience.
  • the present invention provides a method for controlling overload in a P2P network, the method comprising: after a node receiving a message is overloaded, the backup node of the node receiving the message receives a message sent by a node that sends the message.
  • the method further includes:
  • the node that sends the message After the node that sends the message senses that the node that receives the message is overloaded, it sends a message directly to the backup node of the node that receives the message; or
  • the message sent by the node that sends the message is forwarded to the backup node.
  • the method further includes:
  • the backup node After receiving the message, the backup node proxies the node receiving the message to process the service, and after the processing is completed, returns a service request response message to the node that sends the message.
  • the method before the proxying the node receiving the message to process the service, the method further includes:
  • the backup node sends a proxy solicitation request message to the node that receives the message, and after receiving the proxy solicitation request message response message, proxying the node receiving the message to process the service;
  • the backup node sends a proxy solicitation request message to the node that receives the message; after the timer expires, and the message returned by the node receiving the message is not received, the backup node sets the user data to be proxied. Synchronizing the identity, and then proxying the node receiving the message to process the service.
  • One step includes:
  • the node receiving the message After the node receiving the message receives the agent solicitation request message, and allows the backup node agent to process the service, the node receiving the message sets an incompleteness identifier for the user data to be proxied by the backup node, and then The backup node returns a reply message.
  • the incompleteness identifier includes the address information of the backup node; after the incompleteness identifier is set for the user data to be proxied by the backup node, the method further includes:
  • the node receiving the message synchronizes the incompleteness identifier to all backup nodes of the node receiving the message.
  • the method further includes:
  • the node receiving the message performs data synchronization with the backup node.
  • the backup node synchronizes the change data generated when the node receiving the message processes the service to the node receiving the message.
  • the method further includes:
  • the node that sends the message discards or sends the message proportionally according to the overload policy.
  • the method further includes:
  • the node that receives the message is overloaded, and the node that receives the message or the other backup node of the node that receives the message receives the message sent by the node that sends the message again, and then forwards the message to the corresponding user.
  • the address marked in the data area is marked in the data area.
  • the present invention also provides a method for data migration in a P2P network node overload process, Methods include:
  • the new joining node sends a join request message to the responsible node
  • the responsible node migrates the data that the newly added node is responsible for to the newly joined node; after the new joining node receives the migration data, and determines that the data of some users in the received migration data is incomplete, the The backup node responsible for the node acquires data of the part of the user.
  • the determining that the data of the part of the user in the migrated data is incomplete is:
  • the newly joining node determines that the data of the user is incomplete according to the incompleteness identifier in the data of the user.
  • the incompleteness identifier includes the address information of the backup node; the acquiring, by the backup node of the responsible node, the data of the part of the user, where: the new joining node is according to the backup node Address information, obtaining data of the part of the user from the backup node.
  • the data of the part of the user includes the change data generated when the backup node agent.
  • the present invention also provides a system for controlling overload in a peer-to-peer network, the system comprising: a first node, which is a backup node of a node that receives a message, and a node that receives a message is overloaded, and sends a node that sends a message to send Message.
  • the first node is further configured to: after receiving the message sent by the second node, the node that receives the message processing the service, and after the processing is completed, return a service request response message to the second node;
  • the system further includes: a second node, configured to receive a service request response message returned by the second node.
  • the system further includes: a third node, configured to: after receiving the agent solicitation request message sent by the first node, set an incompleteness identifier for the data to be proxied by the first node, Returning a response message to the first node;
  • the first node is further configured to send a proxy solicitation request message to the third node, and after receiving the response message returned by the third node, proxy the third node to process the service; or
  • the first node is further configured to send a proxy solicitation request message to the third node, and after the timer expires, and the message returned by the third node is not received, The data settings are not synchronized, and the third node is acting on the business.
  • the third node after being used for overloading, forwards the received message sent by the second node to the first node;
  • the second node is further configured to send a message to the third node;
  • the second node is further configured to: after sensing that the third node is overloaded, directly send a message to the first node.
  • the third node is further configured to perform data synchronization with the first node after the overload state is canceled;
  • the first node is further configured to perform data synchronization with the third node after the third node is overloaded.
  • the first node is further configured to: after receiving the message sent by the second node, and notify all the backup nodes of the third node that the load is overloaded, notify the second node;
  • the second node is further configured to: after receiving the notification of the first node, discard or send the message according to the overload policy.
  • the present invention also provides a system for data migration during a P2P network node overload process, the system comprising: a fourth node, a fifth node, and a sixth node;
  • the fourth node when used to join the peer-to-peer network, sends a join request message to the fifth node, and after receiving the migrated data migrated by the fifth node, and determining that the data of some users in the received migration data is incomplete, Obtaining data of the part of the user from the sixth node;
  • the fifth node is further configured to: after receiving the join request message sent by the fourth node, migrate the data that the fourth node should be responsible to to the fourth node;
  • the sixth node is configured to provide data of the part of the user to the fourth node.
  • the data of the part of the user includes the change data generated when the sixth node agent.
  • the method and system for controlling overload in a P2P network provided by the present invention, after the node receiving the message is overloaded, the backup node of the node receiving the message receives the message sent by the node sending the message, when the node receiving the message is overloaded, The node receiving the message by the backup node agent of the node receiving the message processes the service, so that when the node of the telecommunication network introducing the P2P network technology is under heavy load, the service connection rate is effectively improved, thereby ensuring the continuity of the telecommunication service, and further Improve user physical insurance.
  • the newly joining node sends a join request message to the responsible node; the responsible node migrates data to be newly added by the newly added node to the newly joined node; After the new joining node receives the migration data, and determines that the data of some users in the received migration data is incomplete, the data of the part of the user is obtained from the backup node of the responsible node, so that the P2P network technology can be introduced.
  • the nodes of the telecommunication network are under heavy load, the integrity of the migrated data is ensured, thereby ensuring the normal operation of the network and improving the user experience.
  • FIG. 1 is a schematic diagram of a telecommunication network architecture of an existing P2P network technology
  • FIG. 2 is a schematic flow chart of a conventional network overload control method
  • FIG. 3 is a schematic flowchart of a method for controlling an overload in a P2P network according to the present invention
  • FIG. 4 is a schematic flowchart of a method for controlling an overload in a P2P network according to Embodiment 1;
  • FIG. 5 is a schematic flowchart of a method for controlling an overload in a P2P network according to Embodiment 2;
  • FIG. 6 is a schematic flowchart of a method for controlling an overload in a P2P network in Embodiment 3; 7 is a schematic flowchart of a method for controlling an overload in a P2P network in Embodiment 4;
  • Embodiment 8 is a schematic flowchart of a method for implementing data migration in Embodiment 5;
  • FIG. 9 is a schematic structural diagram of a system for controlling overload in a P2P network according to the present invention. detailed description
  • the basic idea of the present invention is: After the node receiving the message is overloaded, the backup node of the node receiving the message receives the message sent by the node transmitting the message.
  • the method for controlling the overload in the P2P network of the present invention includes the following steps: Step 301: After the node receiving the message is overloaded, the backup node of the node that receives the message receives the message sent by the node that sends the message;
  • the method may further include:
  • the node that sends the message directly sends the message to the backup node after the node that receives the message is overloaded by the prior art;
  • the message sent by the node that sent the message is forwarded to the backup node.
  • Step 302 After receiving the message, the backup node processes the service of the node receiving the message, and after the processing is completed, returns a service request response message to the node that sends the message; the backup node receives the message and Before the agent receiving the message processes the service, the method may further include:
  • the backup node sends a proxy solicitation request message to the node that receives the message; after receiving the request message, the node receiving the message sets an incompleteness identifier for the user data to be proxied by the backup node, and then goes to the backup.
  • the node returns a reply message;
  • the backup node After receiving the response message, the backup node proxyes the node that receives the message to process the service; or The backup node sends a proxy solicitation request message to the node that receives the message; after the timer expires, and the message returned by the node receiving the message is not received, the backup node sets the user data to be proxied. Synchronizing (Non-SYN) identification, and then proxying the node receiving the message to process the service;
  • the incompleteness identifier is used to identify that the data of the user is incomplete.
  • the incompleteness identifier may further include address information of the backup node, for example, a node (Node) identifier (ID) of the backup node.
  • the IDentity of the information after setting the incompleteness identifier for the user data to be proxied by the backup node, the method may further include: the node receiving the message synchronizing the incompleteness identifier to the received message All backup nodes of the node; then if the message received by the node receiving the message or the other backup node of the node receiving the message is the service of the user of the backup node agent, then forwarding the message to the Describe the address marked in the user data area, that is, the address of the backup node;
  • the non-integrity identifier may be a semi-active identifier; the Semi-Active identifier is used to identify that the user's service has been processed by the backup node proxy; and the Non-SYN identifier is used to indicate that the user's data is not Synchronize;
  • the backup node When the timer expires and the message of the node receiving the message is not received, the backup node considers that the node receiving the message is in a full load state.
  • the present invention also provides a method for data migration in a P2P network node overload process, the method comprising:
  • the new joining node sends a join request message to the responsible node
  • the responsible node migrates the data that the newly added node should be responsible for to the newly joined node.
  • the newly added node After the newly added node receives the migration data, and determines that the data of the part of the received migration data is incomplete, the data of the part of the user is obtained from the backup node of the responsible node;
  • the responsible node is overloaded, and the data that the newly added node should be responsible for is migrated to the new one.
  • the specific processing of the ingress node can adopt the existing processing procedure;
  • Non-Trust non-trusted identifier
  • the newly joined node determines that some of the users in the received migration data are incomplete;
  • the Non-Trust identifier is used to identify that the data is untrustworthy.
  • the node receiving the message cancels its full load state, it sets a Non-Trust identifier for all data maintained by itself.
  • the incompleteness identifier may further include address information of the backup node, such as: node ID information of the backup node; the new joining node acquires the location from the backup node according to the address information of the backup node.
  • address information of the backup node such as: node ID information of the backup node; the new joining node acquires the location from the backup node according to the address information of the backup node.
  • the data of the part of the user includes the change data generated when the backup node agent.
  • the data is synchronized with the backup node. Specifically, the backup node synchronizes the change data generated when the node that receives the message processes the service to the received message. Node.
  • backup node when the backup node receives the message, and all the backup nodes of the node that senses the received message are overloaded, notify the node that sends the message; after the node that sends the message receives the notification, according to the overload policy, Discard or send messages proportionally.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the application scenario of this embodiment is as follows: Node 1 is a node that sends a message, Node 2 is a node that receives a message, and is overloaded, and node 3 is a backup node of node 2.
  • the overload refers to the load of node 2 is reached.
  • the method for controlling the overload in the P2P network in this embodiment includes the following steps:
  • Step 401 After the node 1 senses that the node 2 is overloaded, the node 1 selects to send a service request message to the node 3 according to the overload policy;
  • the node 1 can be aware that the node 2 is overloaded by the prior art; the node 1 can obtain the node 3 through the DHT algorithm, and then send a service request message to the node 3 according to the data;
  • the overload strategy may be set according to requirements, and the overload strategy may include: an overloaded valve value, a full-load valve value, etc.; for example: a valve value that can be set to an overload is 90%, and a full-load valve value is 96%, etc.;
  • the service request message includes: a user identity or a resource ID;
  • the node 1 does not sense that the node 2 is overloaded. After the node 1 sends a service request message to the node, the node 2 forwards the service request message to the node 3 due to overload.
  • Step 402 After receiving the service request message, the node 3 sends a proxy solicitation request to the node 2, where the proxy solicitation request message includes: a user identity identifier or a resource ID.
  • Step 403 Node 2 receives the agent solicitation request message, if the node 3 is allowed to proxy the service, step 404 is performed, otherwise, step 407 is performed;
  • the node 2 determines whether to allow the node 3 to proxy the service according to the need. For example, after the node 2 is overloaded, another backup node of the node 2 proxies the service of a certain user, and the service request received by the node 3 When the information is still the service of the user, node 2 may not allow node 3 to proxy the service, and still another node of node 2 proxy the service of the user.
  • Step 404 The node 2 sets a Semi-Active identifier in the data area corresponding to the user identity or the resource ID, and then returns a response message to the node 3;
  • the Semi-Active identifier may include: address information of the node 3, such as: Node (Node) identification ID information of the node 3; the Semi-Active identifier is used to identify that the service of the user has been processed by the backup node agent;
  • Node 2 can synchronize the set Semi-Active ID to all of its own backup nodes.
  • Step 405 After receiving the response message, the node 3 processes the service instead of the node 2, and returns a service request response message to the node 1 after the processing is completed, and then proceeds to step 406;
  • the proxy node 2 processes the service, or
  • the node 3 can directly process the service by the proxy node 2 according to the protocol with the node 2; if the service request sent by the node 1 only involves the read operation service, the node 3 can not request the consent of the node 2, that is, the step is not executed. 402 ⁇ 404, the direct proxy node 2 processes the service; wherein the protocol can be set according to requirements.
  • Step 406 After the node 2 overload state is canceled, the node 3 performs data synchronization with the node 2, and the node 3 synchronizes the change data generated when the proxy node 2 processes the service to the node 2, and ends the current processing flow;
  • the node 3 may be notified, and after receiving the notification, the node 3 performs data synchronization with the node 2; or, after the node 2 cancels the overload state, it determines whether the Semi-Active identifier exists.
  • User data if present, acquires data from the node 3, thereby implementing data synchronization; or, the node 2 can directly acquire data from the node 3 according to the address information of the node 3 in the Semi-Active identifier, thereby implementing data synchronization; or
  • the node 3 periodically queries the overload state of the node 2, and finds that the node 2 synchronizes the data to the node 2 after the overload state is canceled.
  • the node 3 can query whether the overload state of the node 2 is canceled by periodically sending a message to the node 2.
  • the overload server may be periodically queried whether the overload status of the node 2 is canceled.
  • Step 407 Node 2 returns a reject message to node 3.
  • Step 408 After receiving the reject message, the node 3 returns a reject message to the node 1, and then ends the current processing flow.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the application scenario of this embodiment is: Node 1 is a node that sends a message, Node 2 is a node that receives a message, and is fully loaded, and node 3 is a backup node of node 2.
  • the full load means that the load of node 2 is reached.
  • the preset threshold value of the valve, the method for controlling the overload in the P2P network in this embodiment, as shown in FIG. 5, includes the following steps:
  • Step 501 After node 1 senses that node 2 is fully loaded, node 1 selects according to the overload policy. Sending a service request message to the node 3;
  • the node 1 can be aware that the node 2 is overloaded by the prior art; the node 1 can obtain the node 3 through the DHT algorithm, and then send a service request message to the node 3 according to the data;
  • the overload strategy may be set according to requirements, and the overload strategy may include: an overloaded valve value, a full-load valve value, etc.; for example: a valve value that can be set to be overloaded is 90%, and a full-load valve value is 96%, etc.;
  • the service request message includes: a user identity or a resource ID;
  • the node 1 does not perceive that the node 2 is fully loaded.
  • the node 2 forwards the service request message to the node 3 due to overload.
  • Step 502 After receiving the service request message, the node 3 sends a proxy solicitation request message to the node 2, and starts a timer.
  • the proxy solicitation request message includes: a user identity identifier or a resource ID;
  • the duration of the timer is set according to requirements, such as: Is, etc.
  • Step 503 After the timer expires and the response message of the node 2 is not received, the Non-SYN identifier is set in the data area corresponding to the user identity representation or the resource ID, and then the proxy node 2 processes the service, and after the processing is completed, The node 1 returns a service request response message, and then performs step 504.
  • the node 3 may repeatedly send an acknowledgement message to the node 2 multiple times, if the node is still not received. 2, the confirmation message, the proxy node 2 processes the service; wherein, the number of times the confirmation message is sent can be set according to needs, for example: 3 times;
  • node 3 When the timer expires and the response message of node 2 is not received, node 3 considers node 2 to be in a full load state;
  • the Non-SYN identifier is used to indicate that the data of the user is not synchronized; the Non-SYN identifier may include address information of the node 3, for example: Node ID information of the node 3;
  • step 502 is performed, and After setting the Non-SYN identifier in the data area corresponding to the user identity representation or the resource ID, the proxy node 2 processes the service, or the node 3 can directly process the node 2 according to the protocol with the node 2;
  • the service request sent by the node 1 only involves the read operation service, and the node 3 may not send the agent solicitation request message to the node 2, and the direct proxy node 2 processes the service; wherein the protocol may be set as needed.
  • Step 504 After the node 2 full load state is canceled, the node 3 performs data synchronization with the node 2, synchronizes the change data generated when the proxy node 2 processes the service to the node 2, and ends the current processing flow; where, when the node 2 is fully loaded After canceling, all the data that it maintains is considered untrustworthy. You can set the Non-Tmst identifier for all the data it maintains, and then obtain the change data from the node 3 to achieve data synchronization.
  • the The node 3 queries the data status message, and after receiving the query data status message, the node 3 sends all the user data lists with the Non-SYN identifier to the node 2, and after receiving the list, the node 2 is in the user data list.
  • the user data area is set to the Semi-Active identity, and then the node 3 is notified that the full load state is canceled.
  • the node 3 After receiving the notification, the node 3 performs data synchronization with the node 2; wherein, the Semi-Active identifier is used to identify that the user's service has been represented by the backup node. deal with.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • node 1 is a node that sends a message
  • node 2 is a node that receives a message, and is overloaded
  • node 3 and node 4 are backup nodes of node 2
  • node 3 is also overloaded, here
  • the method of controlling the overload in the P2P network in this embodiment, as shown in FIG. 6, includes the following steps:
  • Step 601 After the node 1 senses that the node 2 is overloaded, the node 1 selects to send a service request message to the node 3 according to the overload policy;
  • the node 1 can be aware that the node 2 is overloaded by the prior art; the node 1 can obtain the node 3 through the DHT algorithm, and then send a service request message to the node 3 according to the data;
  • the overload strategy may be set according to requirements, and the overload strategy may include: an overloaded valve value, a full-load valve value, etc.; for example: a valve value that can be set to be overloaded is 90%, and a full-load valve value is 96%, etc.;
  • the service request message includes: a user identity or a resource ID;
  • the node 1 does not sense that the node 2 is overloaded. After the node 1 sends a service request message to the node, the node 2 forwards the service request message to the node 3 due to overload.
  • Step 602 After receiving the service request message, the node 3 finds that it is overloaded, and forwards the received service request message to the node 4;
  • the node 3 can obtain the node 4 through the DHT algorithm, and then forward the service request message to the node 4 according to this;
  • the service request message includes: a user identity or a resource ID.
  • Step 603 After receiving the service request message, the node 4 sends a proxy solicitation request to the node 2, where the node 4 can also send a proxy solicitation request message to the node 3, and after receiving the message, the node 3 forwards the message to the node 2;
  • the agent solicitation request message includes: a user identity or a resource ID.
  • Step 604 Node 2 receives the proxy solicitation request message, if the node 4 is allowed to proxy the service, step 605 is performed, otherwise, step 608 is performed;
  • the node 2 determines whether to allow the node 4 to proxy the service according to the need. For example, after the node 2 is overloaded, another backup node of the node 2 proxies the service of a certain user, and the service request received by the node 4 When the information is still the service of the user, node 2 may not allow node 4 to proxy the service, and still another node of node 2 shall proxy the user's service.
  • Step 605 The node sets a Semi-Active identifier in the data area corresponding to the user identity or the resource ID, and then returns a response message to the node 4;
  • the node 2 can also return a response message to the node 3, and after receiving the message, the node 3 Node 4 forwards the response message;
  • the Semi-Active identifier may include: address information of the node 4, for example: Node ID information of the node 4; the Semi-Active identifier is used to identify that the service of the user has been processed by the backup node proxy;
  • Node 2 can synchronize the set Semi-Active ID to all of its own backup nodes.
  • Step 606 After receiving the response message, the node 4 processes the service instead of the node 2, and returns a service request response message to the node 1 after the processing is completed, and then proceeds to step 607;
  • step 603-605 needs to be performed, and after the node 4 receives the response message of the node 2, the proxy node 2 processes the service, or the node 4 can cooperate with the node 2 Between the agreement, the direct proxy node 2 processes the service; if the service request sent by the node 1 only involves the read operation service, the node 4 may not require the consent of the node 2, that is, the steps 603 to 605 are not executed, and the direct proxy node 2 processes Service; wherein the protocol can be set as needed.
  • Step 607 After the node 2 overload state is canceled, the node 4 performs data synchronization with the node 2, and the node 4 synchronizes the change data generated when the proxy node 2 processes the service to the node 2, and ends the current processing flow;
  • the node 4 can be notified, and after receiving the notification, the node 4 performs data synchronization with the node 2; or, after the node 2 cancels the overload state, it determines whether the semi-active identifier exists.
  • User data acquires data from the node 4, thereby implementing data synchronization; or, the node 2 can directly obtain data from the node 4 according to the address information of the node 4 in the Semi-Active identifier, thereby implementing data synchronization; or The node 4 periodically queries the overload state of the node 2, and finds that the node 2 synchronizes the data to the node 2 after canceling the overload state; wherein, the node 4 can query whether the overload state of the node is canceled by periodically sending a message to the node, or The overload server can be periodically queried whether the overload status of node 2 is canceled. Step 608: Node 2 returns a reject message to node 4.
  • Step 609 After receiving the reject message, the node 4 returns a reject message to the node 1, and then ends the current processing flow.
  • Embodiment 4 is a diagrammatic representation of Embodiment 4:
  • Node 1 is a node that sends a message
  • Node 2 is a node that receives a message, and is overloaded
  • node 3 is a first backup node of node 2, that is, a preferred backup when node 2 is overloaded. All the backup nodes of the node 2 are overloaded.
  • the overload refers to the preset threshold value of the load of the node.
  • the method for controlling the overload in the P2P network in this embodiment is as shown in FIG. 7 . Includes the following steps:
  • Step 701 After the node 1 senses that the node 2 is overloaded, the node 1 selects to send a service request message to the node 3 according to the overload policy;
  • the node 1 can be aware that the node 2 is overloaded by the prior art; the node 1 can obtain the node 3 through the DHT algorithm, and then send a service request message to the node 3 according to the data;
  • the overload strategy may be set according to requirements, and the overload strategy may include: an overloaded valve value, a full-load valve value, etc.; for example: a valve value that can be set to be overloaded is 90%, and a full-load valve value is 96%, etc.;
  • the service request message includes: a user identity or a resource ID;
  • the node 1 does not sense that the node 2 is overloaded. After the node 1 sends a service request message to the node, the node 2 forwards the service request message to the node 3 due to overload.
  • Step 702 After receiving the service request message, the node 3 senses that all the backup nodes of the node 2 are overloaded, and notifies the node 1;
  • Step 703 After receiving the notification, the node 1 discards or sends the data according to the overload policy.
  • the specific processing procedure for the node 1 to discard or send the message proportionally may be used. Process.
  • Embodiment 5 is a diagrammatic representation of Embodiment 5:
  • the application scenario of this embodiment is as follows: When a new node joins the P2P network, the original responsible node is overloaded, and the original responsible node adopts the control overload method in the first embodiment, the second embodiment, and the third embodiment.
  • the backup node agent responsible for the node is responsible for the node processing service.
  • the method for implementing data migration in this embodiment includes the following steps:
  • Step 801 The newly added node sends a join request message to the original responsible node to obtain the data to be responsible.
  • Step 802 After receiving the request message, the original responsible node migrates the data that the newly added node is responsible for to the newly added node according to the P2P network algorithm.
  • the specific processing process of migrating the data that the newly added node is responsible for to the newly added node is exactly the same as the existing processing, and details are not described herein again.
  • Step 803 The newly joined node receives the migration data, and finds that the change data generated when the agent obtains the proxy from the backup node of the original responsible node after the Semi-Active or Non-Trust identifier is partially set in the migration data;
  • the new joining node finds that the existing part of the mobile data is set with the Semi-Active or Non-Trust identifier, it is considered that the data currently set with the Semi-Active or Non-Tmst identifier is incomplete data, and needs to be from the original responsible node.
  • the change data generated by the backup node when acquiring the proxy; the newly added node may obtain the change data generated by the proxy from the backup node of the original responsible node according to the address information carried in the identifier; or, may obtain the backup node of the original responsible node according to the DHT algorithm. According to this, the change data generated when the proxy is obtained from the backup node of the original responsible node.
  • Step 804 The backup node of the original responsible node migrates the change data generated when the agent is responsible for processing the service to the newly joined node.
  • the present invention further provides a system for controlling overload in a P2P network.
  • the system includes: a first node 91, and a Two nodes 92; wherein
  • the first node 91 after the node for receiving the message is overloaded, receives the message sent by the second node 91;
  • the second node 92 after sensing that the node receiving the message is overloaded, directly sends a message to the first node 91.
  • the first node 91 is a backup node of a node that receives a message
  • the second node 92 is a node that transmits a message.
  • the first node 91 is further configured to: after receiving the message sent by the second node 92, proxy the node receiving the message to process the service, and return the service request response message to the second node 92 after the processing is completed;
  • the second node 92 is further configured to receive a service request response message returned by the first node 91.
  • the system may further include: a third node 93, configured to: after receiving the proxy solicitation request message sent by the first node 91, set an incompleteness identifier for the data to be proxied by the first node 91, and then return to the first node 91 Reply message
  • the first node 91 is further configured to send a proxy solicitation request message to the third node 93, and after receiving the response message returned by the third node 93, proxy the third node 93. Processing business; or,
  • the first node 91 is further configured to send a proxy solicitation request message to the third node 93, and after the timer expires, and no message returned by the third node 93 is received, , setting Non-SYN for the data to be proxied, and proxying the third node 93 to process the service;
  • the third node 93 refers to a node that receives a message
  • the first node 91 is a backup node of the third node 93.
  • the third node 93 is further configured to forward the received message sent by the second node 92 to the first node 91;
  • the second node 92 is further configured to send a message to the third node 93.
  • the third node 93 is further configured to perform data synchronization with the first node 91 after the overload state is canceled;
  • the first node 91 is further configured to perform data synchronization with the third node 93 after the third node 93 cancels the overload state.
  • the first node 91 is further configured to notify the second node 92 after receiving the message sent by the second node 92 and sensing that all the backup nodes of the third node 93 are overloaded;
  • the second node 92 is further configured to: after receiving the notification of the first node 91, discard or send the message proportionally according to the overload policy.
  • the present invention further provides a system for data migration in a process of overloading a P2P network node based on the method for data migration in a P2P network node overload process, the system comprising: a fourth node, a fifth node, and a sixth node; among them,
  • the fourth node when used to join the peer-to-peer network, sends a join request message to the fifth node, and after receiving the migrated data migrated by the fifth node, and determining that the data of some users in the received migration data is incomplete, Obtaining data of the part of the user from the sixth node;
  • the fifth node is further configured to: after receiving the join request message sent by the fourth node, migrate the data that the fourth node should be responsible to to the fourth node;
  • the sixth node is configured to provide data of the part of the user to the fourth node.
  • the fifth node is overloaded
  • the sixth node is the backup node of the fifth node.
  • the data of the part of the user includes the change data generated when the sixth node agent is used.
  • the specific processing procedure of the fourth node and the sixth node in the system for data migration in the overload process of the P2P network node of the present invention has been detailed in the above, and will not be described again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

L'invention porte sur un procédé de maîtrise de surcharge dans un réseau poste à poste, comprenant l'opération suivante : une fois qu'un nœud de réception de message est surchargé, un nœud de secours du nœud de réception de message reçoit un message envoyé par un nœud d'envoi de message. L'invention porte également sur un système de maîtrise de surcharge dans un réseau poste à poste. Avec le procédé et le système de la présente invention, lorsqu'un nœud d'un réseau de télécommunication dans lequel la technologie de réseau poste à poste est introduite est lourdement chargé, le taux de connexion au service peut être efficacement amélioré, ce qui assure la continuité de services de télécommunication et améliore l'expérience utilisateur.
PCT/CN2012/072273 2011-04-08 2012-03-13 Procédé et système de maîtrise de surcharge dans un réseau poste à poste WO2012136094A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110087745.2 2011-04-08
CN201110087745.2A CN102739711B (zh) 2011-04-08 2011-04-08 一种对等网络中控制过负荷的方法及系统

Publications (1)

Publication Number Publication Date
WO2012136094A1 true WO2012136094A1 (fr) 2012-10-11

Family

ID=46968582

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/072273 WO2012136094A1 (fr) 2011-04-08 2012-03-13 Procédé et système de maîtrise de surcharge dans un réseau poste à poste

Country Status (2)

Country Link
CN (2) CN106850730A (fr)
WO (1) WO2012136094A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103036928A (zh) * 2011-10-10 2013-04-10 中兴通讯股份有限公司 网络过载的处理方法及装置
CN110658779A (zh) * 2019-10-09 2020-01-07 翼石电子股份有限公司 一种plc数据存储方法及装置

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101399765A (zh) * 2007-09-28 2009-04-01 华为技术有限公司 降低对等网络中热点节点负荷的方法和系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3973548B2 (ja) * 2002-12-10 2007-09-12 株式会社ソニー・コンピュータエンタテインメント ネットワークシステム、ネットワーク接続確立方法、ネットワーク端末、コンピュータプログラム、およびプログラムを格納した記録媒体
EP1713233A1 (fr) * 2005-04-14 2006-10-18 Alcatel Interconnexion des domaines dans un réseau entre homologues
CN101242251A (zh) * 2007-02-07 2008-08-13 华为技术有限公司 提高p2p重叠网稳定性的方法和装置
CN101296108B (zh) * 2007-04-27 2012-12-12 华为技术有限公司 一种在结构化p2p中备份资源的方法和系统
CN101399746B (zh) * 2007-09-26 2011-03-16 华为技术有限公司 报文路由方法、系统、设备和选择备份资源的方法、系统
CN101729277B (zh) * 2008-10-27 2015-07-08 华为技术有限公司 设备池的管理方法、节点设备和通信系统

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101399765A (zh) * 2007-09-28 2009-04-01 华为技术有限公司 降低对等网络中热点节点负荷的方法和系统

Also Published As

Publication number Publication date
CN102739711B (zh) 2017-10-17
CN102739711A (zh) 2012-10-17
CN106850730A (zh) 2017-06-13

Similar Documents

Publication Publication Date Title
JP6047229B2 (ja) 情報中心ネットワークにおける名前ベースの近隣探索及びマルチホップサービス探索
JP5847191B2 (ja) コンテンツ共有を行う中間ノード及びコンテンツリクエスト端末並びにそれらのコンテンツ共有方法
US8380854B2 (en) Simplified method for processing multiple connections from the same client
US20100223380A1 (en) Session Monitoring Method, Apparatus, and System Based on Multicast Technologies
EP2171600A2 (fr) Transmission multimédia en continu poste-à-poste assistée
US20120077483A1 (en) Enabling interface aggregation of mobile broadband network interfaces
WO2012048613A1 (fr) Système d'homologue à homologue (p2p) centralisé qui supporte une traversée nat et procédé de traversée nat
WO2013040970A1 (fr) Procédé et dispositif de sélection de nœud de relais
WO2012065531A1 (fr) Procédé, dispositif, et système pour effectuer une sélection de relais
WO2013143360A1 (fr) Procédé de téléchargement de données de service et terminal mobile
US9350652B2 (en) Method, system and router for changing application in BGP session
US11134001B2 (en) Data distribution method and distribution server
WO2009000181A1 (fr) Procédé servant à acquérir des ressources transversales, un nœud pair à pair et un système pair à pair
WO2012167659A1 (fr) Procédé et dispositif de communication de données dans un protocole d'applications contraintes
CN104660550A (zh) 一种在多服务器之间进行会话迁移的方法
WO2012136094A1 (fr) Procédé et système de maîtrise de surcharge dans un réseau poste à poste
WO2013053252A1 (fr) Procédé et appareil de traitement des surcharges de réseau
WO2012100560A1 (fr) Procédé et nœuds pour mettre en œuvre une migration de données au cours d'une session
JP4800916B2 (ja) データ中継装置
JP3682439B2 (ja) データ通信システム及び方法、サーバ装置、クライアント装置、並びにプログラム
WO2012167710A1 (fr) Procédé et système de sauvegarde de services dans un réseau à nœuds hétérogènes
KR102090493B1 (ko) 무선 통신 네트워크에서 http 프로토콜의 전송 지연과 http 서버의 프로세싱 부하를 줄이는 장치 및 방법
EP2790362B1 (fr) Procédé pour la régulation d'une charge sur un réseau et serveur d'enregistrement
WO2012129794A1 (fr) Procédé de communication, nœud de réseau et super-nœud de réseau dans un réseau d'homologues (p2p)
WO2012048576A1 (fr) Procédé, serveur et système de commande de téléchargement p2p (poste à poste)

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12768255

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12768255

Country of ref document: EP

Kind code of ref document: A1