WO2013060221A1 - Procédé d'addition de nœuds pour système p2p, système p2p correspondant, procédé et système de rétrogradation de nœuds d'un réseau homologue, et nœud de réseau homologue - Google Patents

Procédé d'addition de nœuds pour système p2p, système p2p correspondant, procédé et système de rétrogradation de nœuds d'un réseau homologue, et nœud de réseau homologue Download PDF

Info

Publication number
WO2013060221A1
WO2013060221A1 PCT/CN2012/082415 CN2012082415W WO2013060221A1 WO 2013060221 A1 WO2013060221 A1 WO 2013060221A1 CN 2012082415 W CN2012082415 W CN 2012082415W WO 2013060221 A1 WO2013060221 A1 WO 2013060221A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
peer
client
candidate
successor
Prior art date
Application number
PCT/CN2012/082415
Other languages
English (en)
Chinese (zh)
Inventor
邓灵莉
彭晋
马啸
Original Assignee
中国移动通信集团公司
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
Priority claimed from CN201110334787.1A external-priority patent/CN103095744B/zh
Priority claimed from CN201110333816.2A external-priority patent/CN103095742B/zh
Application filed by 中国移动通信集团公司 filed Critical 中国移动通信集团公司
Publication of WO2013060221A1 publication Critical patent/WO2013060221A1/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
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks

Definitions

  • Node joining method for P2P system and corresponding P2P system, peer-to-peer network node degradation method and system, peer-to-peer network node The application request is submitted to the Chinese Patent Office on October 28, 2011, the application number is 201110333816.2, and the invention name is "Public joining method for P2P system and corresponding P2P system, Chinese patent application, and submitted to China Patent Office on October 28, 2011, application number 201110334787.1, the invention name is "peer-to-peer network node degradation method and The priority of the Chinese Patent Application for System, Peer-to-Peer Networking, the entire contents of which is incorporated herein by reference.
  • the present invention relates to the field of peer-to-peer network technologies, and in particular, to a node joining method for a P2P system, a corresponding P2P system, a peer-to-peer network node degradation method and system, and a peer-to-peer network node.
  • Background technique
  • RELOAD Resource Location and Discovery Base Protocol
  • Nodes are divided into two categories: strong performance, environment Stable and willing to serve other people as peer nodes, participating in message routing and distributed network services such as distributed data storage; other subjective or objective reasons are not suitable for participating in service provision as a client node (client).
  • a node has its own independent and unique identifier in the peer-to-peer network, that is, the node identifies the Node-ID.
  • Peer resource ID Resource-ID Peer resource ID Resource-ID.
  • Peer-to-peer networks have good self-organization and self-management features, allowing peer nodes to freely join and exit the network.
  • RELOAD introduces a centralized registration server (ES) control node (including peer nodes and client nodes), and assigns a Node-ID to the node to which the application is applied and assigns a secure boot to its initial access network.
  • ES centralized registration server
  • BP Bootstrap Peer
  • load balancing is one of the key factors to ensure the quality of peer-to-peer network services.
  • the structured peer-to-peer network uses the hash method to determine the storage and service location of each data object, so that the load imbalance will inevitably occur.
  • the local load in the peer-to-peer network is too heavy, and the corresponding peer node becomes a local overload node.
  • Overloading of local service nodes directly affects the user experience of the corresponding users, and in some extreme cases triggers a chain reaction that causes the network to collapse (ie, the avalanche effect).
  • RELOAD is a unified P2P protocol between the client node and the peer node.
  • the client node does not participate in message routing when joining the network.
  • the Node-ID statically assigned by the ES is later upgraded to the peer node, the actual decision is made.
  • RELOAD recommends a randomly generated method. Under the premise that the performance of each node is not much different and the spatial distribution of the Node-ID generated by the random algorithm is relatively uniform, this design can play a role of static load balancing to a certain extent.
  • an embodiment of the present application discloses a P2P node access method, including:
  • the overload peer node or management server obtains real-time performance values of candidate client nodes directly connected to the overload peer node;
  • a P2P system including: An overload peer node or a management server obtains a real-time performance value of a candidate client node directly connected to the overload peer node, and determines a candidate client node whose real-time performance value is greater than a preset performance threshold as a client node to be upgraded, and select at least A client node to be upgraded and upgraded to a peer node.
  • the information of the directly connected client node (especially the candidate client node) is locally collected by the overload peer node, and the load is loaded according to the desired performance, and the appropriate client node is selected to be reassigned according to the performance of the client node itself.
  • the Node-ID is upgraded to a peer node, which avoids the load splitting failure caused by the client node resource and environment constraints and cannot bear the corresponding Overlay workload.
  • an embodiment of the present application discloses a peer to peer network node degradation method, the method comprising: sending a leave request message to a successor node of a peer to peer network in a peer to peer network; The leaving the consent message sent by the successor node, the peer node to be demoted degraded itself to the client node, wherein the leaving consent message is that the successor node determines that the peer node to be demoted can be processed according to its current processing capability. After the business is sent.
  • a peer-to-peer network node being a peer node to be demoted, comprising: a leave request message sending unit, configured to send a leave request message to a successor node of the peer network node; a consent message receiving unit, configured to receive a leave consent message sent by the successor node; a downgrade unit, configured to: after the leave consent message receiving unit receives the leave consent message, demote the peer network node by a peer node And being the client node, where the leaving consent message is sent by the successor node after determining that the service of the peer network node can be processed according to its current processing capability.
  • Another embodiment of the present application discloses a peer-to-peer network node degradation method, where the method includes: after a successor node of a peer to peer network in a peer-to-peer network receives a leave request message sent by the peer node to be demoted, according to Determining whether the service of the peer node to be demoted can be processed by the current processing capability; if the determination result is yes, the successor node sends a leave consent message to the to-be-degraded peer node, indicating the to-be-degraded pair The nodes are downgraded to client nodes.
  • a peer-to-peer network node is further disclosed, where the peer-to-peer network node is a successor node of the peer node to be demoted, and includes: a leave request message receiving unit, configured to receive the leave request sent by the peer node to be demoted a judging unit, configured to determine, according to the current processing capability of the peer network node, whether the service of the peer node to be demoted can be processed; the second leaving consent message sending unit, And when the determining result of the determining unit is yes, sending a leaving consent message to the to-be-degraded peer node, indicating that the to-be-degraded peer node is demoted to a client node.
  • a peer-to-peer network node degradation system including a peer node to be demoted in a peer-to-peer network and a successor node of the peer node to be demoted, wherein: a peer node to be demoted, for sending to its successor node The leaving request message, if receiving the leaving consent message sent by the successor node, degrading itself to the client node; the successor node, after receiving the leaving request message sent by the to-be-degraded peer node, according to its current processing capability And determining whether the service of the peer node to be demoted can be processed, and if the determination result is yes, sending a leave consent message to the to-be-degraded peer node.
  • the peer node to be demoted in the peer-to-peer network sends a leave request message to its successor node, and the successor node determines whether the service of the peer node to be demoted can be processed according to its current processing capability. If it can be processed, the departure consent message is sent to the to-be-degraded peer node, and the degraded peer node degrades itself to the client node only after receiving the leave consent message, as can be seen from the above, the peer to peer network is to be demoted. The node no longer sends the leave request message to the successor node and then directly degrades and then exits the network. Instead, the node is downgraded to the client node after the successor node agrees.
  • the peer node to be demoted can only process the subsequent node.
  • the service of the peer node is degraded, which effectively avoids the situation that the peer node in the peer-to-peer network is directly downgraded to the client node, causing the load of its successor node to be too heavy, or even causing the peer network to be paralyzed.
  • FIG. 1 is a schematic diagram of a P2P system according to an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a P2P node access method 1000 according to an embodiment of the present application
  • FIG. 3 is a schematic flowchart of a P2P node access method 2000 according to an embodiment of the present application
  • FIG. 4 is a flow of step S260 of FIG. Schematic diagram
  • FIG. 5 is a schematic flow chart of step S280 in Figure 2;
  • FIG. 6 is a schematic flowchart of revocation of a candidate upgrade node according to an embodiment of the present application
  • FIG. 7 is a schematic diagram of a peer-to-peer network architecture according to Embodiment 1 of the present invention
  • FIG. 8 is a flow chart of a peer-to-peer network node degradation method on a peer node side to be demoted in a second embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a peer-to-peer network node according to Embodiment 3 of the present invention.
  • FIG. 10 is a schematic flowchart of a peer-to-peer network node degradation method on a successor node side according to Embodiment 4 of the present invention.
  • FIG. 11 is a schematic structural diagram of a peer-to-peer network node according to Embodiment 5 of the present invention.
  • FIG. 12 is a schematic flowchart of a method for degrading a peer-to-peer network node when actively degrading according to Embodiment 6 of the present invention.
  • FIG. 13 is a schematic flowchart of a method for degrading a peer-to-peer network node during passive degrading according to Embodiment 7 of the present invention
  • FIG. 14 is a schematic structural diagram of a peer-to-peer network node degradation system according to Embodiment 8 of the present invention. detailed description
  • FIG. 1 is a schematic diagram of a P2P system in accordance with an embodiment of the present application.
  • the P2P system of this embodiment includes a peer node 11, client nodes 12 and 13 directly connected to the peer node, a management server 21, a configuration server 22, and a registration server 23.
  • 2 is a flow chart of a P2P node access method 1000 according to an embodiment of the present application. The method 1000 illustrated in FIG. 2 is described below in conjunction with the system illustrated in FIG. 1, and assumes that both client nodes 12 and 13 belong to candidate client nodes.
  • the overload peer node (ie, when the peer node 11 is overloaded) obtains the real-time performance value of the candidate client node directly connected thereto.
  • the overload peer node 11 obtains real-time performance values from the client nodes 12 and 13 that are directly connected thereto.
  • the overload peer node 11 compares the acquired real-time performance values of the candidate client nodes 12 and 13 with a preset performance threshold, and determines a candidate client node whose real-time performance value is greater than the preset performance threshold as the to-be-upgraded
  • the nodes for example, the real-time performance values of the candidate client nodes 12 and 13 are all greater than a preset threshold, and the candidate client nodes 12 and 13 are all nodes to be upgraded.
  • step S130 at least one node to be upgraded is upgraded to a peer node.
  • the overload peer node 11 selects a candidate client node with a suitable real-time performance value to upgrade to a peer node according to the load condition that needs to be shunted by itself.
  • an overload notification may be sent to the management server 21, and the management server 21 responds to the overload notification to the candidate client directly connected to the overload peer node 11.
  • Nodes e.g., candidate client nodes 12 and 13) collect real-time performance values.
  • step S120 the management server 21 compares the obtained real-time performance values of the candidate client nodes 12 and 13 with a preset performance threshold, and determines a candidate client node whose real-time performance value is greater than the preset performance threshold as the node to be upgraded. For example, if the real-time performance values of the candidate client nodes 12 and 13 are both greater than a preset threshold, then the candidate client nodes 12 and 13 are nodes to be upgraded. Then in step S130, the management server selects at least one node to be upgraded and upgrades to a peer node.
  • FIG. 3 is a schematic flowchart of a P2P node access method 2000 according to an embodiment of the present application. The method 2000 shown in FIG. 3 will be described below in conjunction with the P2P system shown in FIG.
  • the client node e.g., client nodes 12 and 13 directly connected to the peer node 11 downloads the performance test tool.
  • client nodes 12 and 13 obtain a configuration file from configuration server 22 when joining the network, including the download address and/or file fingerprint of the performance test tool.
  • a configuration file For example, to extend the RELOAD configuration file, add a "benchmark" field to save the download address of the '1" productivity tester, benchmark.url and/or benchmark. checksum.
  • Client nodes 12 and 13 download from the download address specified in it. Performance testing tool.
  • the client nodes 12 and 13 can also calculate the hash fingerprint of the downloaded configuration file by using a specified algorithm, and compare it with the file fingerprint saved in the configuration file. If they are consistent, the download is successful; otherwise, A download error is reported to the configuration server 22.
  • the client node runs a performance test tool to obtain performance test results.
  • the performance test tool may be run locally before the client nodes 12 and 13 contact the registration server 23 to obtain the network access authorization, and the performance test results may be saved to send test results when the candidate client node is further applied, or to overload the peer node or
  • the management server collects the real-time performance values of the directly connected candidate client nodes as a reference.
  • step S230 the client node sends the performance test result and the request upgrade request to the registration server.
  • the registration server For example, suppose customer nodes 12 and 13 voluntarily upgrade to peer nodes when needed, for overload The peer node 11 offloads the load and provides services to other client nodes, and the client nodes 12 and 13 send performance test results, request upgrade requests, and other necessary identity information (such as user names, etc.) to the registration server 23, indicating that they are willing to become candidates. node.
  • step S240 the registration server 23 determines the client node whose performance test result is greater than the preset performance threshold as the candidate client node. For example, the registration server 23 verifies the identity of the user and determines a client node whose performance value is greater than the performance threshold as a candidate client node based on a preset performance threshold.
  • a node certificate can include a username, a node ID, a Node-ID, whether an upgrade is allowed, and other information, and is signed by the CA with its private key.
  • the registration server 23 issues the candidate node certificate to the determined candidate client node, the following steps are also included:
  • the candidate client node sends the node certificate to the peer node responsible for storing the node certificate (referred to as the responsible storage peer node) for saving.
  • the responsible storage peer node responsible for storing the peer node to confirm whether the node certificate is responsible for saving and is within the validity period. If not, discard the storage request.
  • it is responsible for storing the validity of the peer node using the CA public key to verify the signature. Then, the responsible peer node saves the verified node certificate and returns a save success message.
  • each candidate client node receives a performance information acquisition request sent by an overload peer node or a management server.
  • client nodes 12 and 13 receive real-time performance value acquisition requests sent by overload peer node 11 or management server 21.
  • the overload peer node 11 or management server 21 sends a ProbeReq to the client nodes 12 and 13 that are directly connected to the overload peer node 11 to query the real-time performance values of the client nodes 12 and 13.
  • ProbeReq in RELOAD defines three data types that can be queried, including respondable-set, num_resource, and uptime.
  • a new variable client-capability can be defined to represent the real-time performance value of the client node.
  • Client nodes 12 and 13 receive ProbeReq to write real-time performance values to ProbeAns based on actual conditions and then return to overload peer node 11 or management server 21.
  • the candidate client node collects local real-time performance versus data in response to the real-time performance value acquisition request. For example, the candidate client nodes 12 and 13 obtain information such as the real-time usage ratio of the local CPU using an interface provided by the operating system.
  • the candidate client node converts the real-time performance data into a real-time performance value based on the performance test result.
  • candidate client nodes 12 and 13 refer to performance test results to convert real-time performance versus data into real-time performance values (eg, real-time performance values may be reflected as locally measured CPU full load performance of the candidate client node and real-time idleness of the local CPU) The ratio is multiplied, wherein the sum of the real-time idle ratio of the CPU and the above-mentioned CPU real-time usage ratio is 1).
  • the candidate client node returns a real-time performance value.
  • candidate client nodes 12 and 13 return local real-time performance values to overload peer node 11 or management server 21 in accordance with the local resource sharing policy.
  • candidate client nodes 12 and 13 report the local shared processing performance upper limit and the calculated minimum value of the real-time performance value.
  • step S270 the overload peer node 11 or the management server 21 compares the acquired real-time performance values of the candidate client nodes 12 and 13 with a preset performance threshold, and selects a candidate client node whose real-time performance value is greater than a preset performance threshold. Determining the nodes to be upgraded, for example, the real-time performance values of the candidate client nodes 12 and 13 are greater than a preset threshold, then the candidate client nodes 12 and 13 are nodes to be upgraded.
  • step S280 at least one node to be upgraded is upgraded to a peer node.
  • the overload peer node 11 selects a candidate client node whose real-time performance value is suitable to be upgraded to a peer node according to the load condition that is required to be shunted by itself; or the management server 21 selects a real-time according to the load condition of the overloaded peer node 11 to be shunted.
  • a candidate client node with a suitable performance value is upgraded to a peer node.
  • FIG. 5 is a schematic flow chart of step S280 in FIG. 3.
  • the client node to be upgraded is determined.
  • the overload peer node 11 or the management server 21 may verify the node credentials of the candidate client nodes 12 and 13 (eg, the node certificate may be sent to the overload peer node 11 or the management server 21 along with the real-time performance value) to confirm that it has candidates Upgrade qualification; then determine the customer node to be upgraded according to the load that the overload peer node 11 needs to offload and the load that the real-time performance values of the candidate client nodes 12 and 13 can bear, for example, determine the candidate client node 12 as the node to be upgraded; Determines the range of peer-IDs that can be used after upgrading to a peer node for the client node to be upgraded.
  • the overload peer node 11 or the management server 21 sends a rise to the node to be upgraded. Level request.
  • the overload peer node 11 or the management server 21 sends an UpdateReq message to the client node 12 to be upgraded.
  • the UpdateReq message may include: a Note-ID of the client node 12 to be upgraded to indicate that the client node that needs to be upgraded is the client node 12; a range of eer-IDs that can be used after the client node 12 to be upgraded is upgraded to the peer node, , expect-peer-id-top and expect-peer-id- floor; overload peer node 11 ID; node certificate and other information.
  • the node 12 to be upgraded that receives the UpdateReq may return a reply message to the overload peer node 11 or the management server 21 to indicate that the upgrade is in progress.
  • step S2803 the client node to be upgraded forwards the upgrade request to the registration server.
  • the client node 12 to be upgraded sends an http message to the registration server 23, which may include an UpdateReq message sent by the overload peer node 11 or the management server 21 to the client node 12 to be upgraded.
  • the registration server returns an upgrade node certificate.
  • the registration server 23 returns a certificate update node (i.e. the new node certificate)
  • the upgrade certificate node may include specific p eer -lD (i.e., upgrade client node 12 to be upgraded to a peer node 12 to be upgraded particular use peer-ID) and location information of the boot node.
  • step S2805 the node to be upgraded re-accesses the network and is upgraded to a peer node.
  • the node to be upgraded 12 rejoins the network in accordance with the standard procedure of RELOAD with the specific eer-ID returned by the registration server 23.
  • the method and system disclosed by the foregoing embodiments locally collect the real-time performance of the directly connected client node (especially the candidate client node) by the overload peer node in order to avoid the problem of poor or even invalid upgrade caused by statically assigning the Node-ID. Value, according to the expected offloading load of the overload peer node, combined with the performance of the client node itself, select the appropriate client node to reassign the Node-ID and then upgrade to the peer node.
  • test performance result is obtained through a unified performance testing tool when the client node joins the network, and the overload peer node collects the client node.
  • the test performance results can be used to estimate the shunt load that can be tolerated after the upgrade, as the basis for the allocation of the upgraded Node-ID.
  • an upgrade candidate authorization process independent of the original customer node network access authorization is added, and the delegation authorization certificate signed by the overload peer node (or management server) is used. (Whether the upgrade flag is allowed), for third party festivals Verify the legality of the customer upgrade request.
  • FIG. 6 is a schematic flowchart of revoking a candidate upgrade node according to an embodiment of the present application.
  • the overload peer node or the management server can also monitor the determined candidate client node. For example, when the overload peer node 11 or the management server 21 determines the candidate client nodes 12 and 13, the candidate client nodes 12 and 13 remain monitored, for example, monitoring user behavior. As shown in FIG. 6, in step S310, the overload peer node or the management server determines the candidate client node to be sold according to the monitoring information. For example, when the overloaded peer node 11 or the management server 21 finds that the candidate client node 12 frequently goes online and offline according to the monitoring information, it determines it as a candidate client node.
  • step S320 the overload peer node or the management server issues a purchase request to the registration server.
  • the overload peer node 11 or the management server 21 sends 4 sales candidate upgrade requests to the registration server 23, requesting that the candidate client node 12 be revoked.
  • step S330 the registration server issues a candidate upgrade node certificate to the candidate customer node.
  • the registration server 23 issues a revocation candidate upgrade node certificate to the to-be-sell candidate client node 12, and the revocation candidate upgrade node certificate includes a flag of "revocation candidate upgrade”.
  • the candidate client node forwards the candidate upgrade node certificate to the responsible storage node to store the candidate upgrade node certificate.
  • the candidate client node 12 sends the four candidate candidate node certificates to the peer node responsible for storing the node certificate (referred to as the responsible storage node for short).
  • the responsible storage node responsible for storing the peer node to confirm whether the node certificate is responsible for saving and is within the validity period. If not, discard the storage request.
  • it is responsible for storing the validity of the peer node using the CA public key to verify the signature.
  • it is responsible for storing the peer node to locally delete the revoked node certificate, and saving the verified undo candidate node certificate, and returning the save success message.
  • the candidate node certificate After issuing the candidate node certificate to the candidate client node, it may further determine whether the sold candidate client node has been upgraded to the peer node. When the revoked candidate client node has been upgraded to a peer node, the peer node is demoted to a client node.
  • the overload peer node 11 or the management server 21 determines that the candidate upgrade node 12 that has been upgraded has been upgraded to a peer node
  • the overload peer node 11 or the management server 21 or the storage peer node is located and The set of online peer nodes directly connected by the candidate upgrade node 12; the next set of online peer nodes (which may also include sending to the configuration server and/or the public boot node as an option)
  • the candidate of the pin upgrades the revocation notification of the node; then the peer node receiving the revocation notification disconnects the connection with the revoked candidate upgrade node. This causes the peer node to which the candidate client node 12 to be rescinded is upgraded to be forced down to the client node.
  • Peer-to-Peer also known as peer-to-peer network technology, which is used between different nodes without relaying.
  • P2P Peer-to-Peer
  • a peer-to-peer network contains multiple peer-to-peer network nodes. There are two types of peer-to-peer network nodes: peer nodes and client nodes.
  • the resources used for switching in the peer-to-peer network are distributed and stored in each peer node, and each peer node has peering.
  • the peer node can provide services to other nodes (that is, the pair is correct).
  • Other nodes in the network provide the exchange resources stored by themselves, and may also request services from other peer nodes (that is, obtain exchange resources from other peer nodes).
  • the client node can only request services from the peer node (that is, obtain the exchange resources from the peer node).
  • a client node in a peer-to-peer network is a client device, and a peer node is a server.
  • customer nodes with strong capabilities and stable environments in peer-to-peer networks can be upgraded to peer nodes to serve other nodes.
  • the peer node upgraded by the client node to the peer node cannot continue to provide services to other nodes when its own processing resources are insufficient or want to leave the peer network.
  • the peer node can directly Each subsequent node sends a leaving request, and the service of the peer node (that is, the service request to be processed by the peer node) is subsequently processed by each subsequent node, which is equivalent to downgrading the peer node to a client node, the pair
  • the peer node can exit the peer-to-peer network directly after sending the leave request. It can be seen from the above that when the peer node cannot provide services for other nodes, it can be directly downgraded to the client node and exit the peer-to-peer network after sending the leaving request.
  • the peer node Since the peer node is directly downgraded to the client node, the peer node is originally processed by the peer node. The service will be automatically routed to each subsequent node of the peer node for processing, which results in an increase in the load of the successor node of the peer node. If a successor node is overloaded, resulting in insufficient processing resources, the successor node is also If the other nodes cannot be served, then the corresponding service will be routed to each subsequent node of the successor node. A similar phenomenon follows the subsequent relationship between the peer nodes, and the unprocessed service proceeds. It will roll up like a "snowball" and eventually lead to a paralysis of peer-to-peer networks.
  • the embodiment of the present invention further provides a peer-to-peer network node degradation method and system, and a peer-to-peer network node, which are used to solve the problem that the peer node in the peer-to-peer network is directly downgraded to a client node, causing the load of the successor node to be excessive.
  • a peer-to-peer network node degradation method and system and a peer-to-peer network node
  • FIG. 7 it is a schematic diagram of a peer-to-peer network architecture in the first embodiment of the present invention.
  • the topology of each peer node in the peer-to-peer network is connected in a ring shape, and each peer node is small according to the number in the node identifier.
  • the order to the big is clockwise on the ring.
  • Each peer node has at least one predecessor node and at least one successor node in the peer-to-peer network.
  • the peer nodes arranged in a clockwise manner may be ranked in the peer node.
  • the peer node adjacent to the node and adjacent to the peer node is determined as the precursor node of the peer node, and correspondingly, when determining the successor node of the peer node, the peer nodes may be arranged in a clockwise manner
  • a peer node that is ranked after the peer node and that is adjacent to the peer node is determined to be a successor node of the peer node.
  • the predecessor node of peer node N1 is peer node N56
  • its successor node is peer node N8.
  • the exchange resource in the peer-to-peer network has a corresponding resource identifier, and the peer node that stores the exchange resource is determined according to the preset algorithm and the resource identifier of the exchange resource.
  • the preset algorithm used in the first embodiment of the present invention is a distribution.
  • the DHT (Dash) algorithm calculates the resource identifier according to the DHT algorithm, and obtains the node identifier of the peer node storing the exchange resource.
  • the storage resource identifier is K10.
  • the peer node is the peer node N14.
  • the peer-to-peer network node degrading method provided by the embodiment of the present invention includes a degrading method on the peer node side to be demoted and a degrading method on the succeeding node side.
  • the degrading method on the peer node side to be demoted is first introduced.
  • FIG. 8 it is a schematic flowchart of a method for degrading a peer-to-peer network node on a peer node side to be demoted according to Embodiment 2 of the present invention, and the specific processing procedure is as follows:
  • Step S410 The peer node to be demoted in the peer-to-peer network sends a leave request message to its successor node, where the peer node to be demoted is a peer node served by the client device, and the client device is upgraded to peer by the client node.
  • the peer node is referred to as a peer node to be demoted;
  • the leave request message is used by the peer to be degraded to indicate to its successor node that it has a downgrade, and asks whether its successor node agrees with its own downgrade request.
  • the peer node to be demoted may only send a leave request message to the succeeding node; the peer node in the peer-to-peer network may also correspond to multiple successor nodes, if If the degraded peer node has multiple successor nodes, the to-be-degraded peer node may separately send a leave request message to each subsequent node.
  • the degraded peer nodes in the peer-to-peer network can be classified into active degraded and passive degraded.
  • Active degrading if the peer node in the peer-to-peer network is served by the client device, when the peer node has insufficient processing resources or the user experience is not good, the user may actively perform the downgrade, that is, actively to the peer node.
  • the successor node sends a leave request message;
  • a decision node in a peer-to-peer network learns that the routing delay of the peer network is too large or other anomalies, it is considered necessary to have some pairs that are served by the client device.
  • the nodes are downgraded to client nodes, and no longer serve other nodes to alleviate the excessive routing delay.
  • the decision node can first perform real-time status query on each peer node in the peer-to-peer network that is served by the client device.
  • each peer node to be demoted determines, according to the real-time status of the query, each peer node to be demoted, and then sending a downgrade request message to each determined peer node to be demoted, after the degraded peer node receives the downgrade request message sent by the decision node, Confirm that it is going to downgrade to a client node and then send a leave request message to its successor.
  • the real-time status is a CPU (Central Processing Unit) usage ratio
  • the decision node first queries the CPU usage ratio of each peer node served by the client device, and then uses the CPU usage ratio below a predetermined value. Peer section The point is determined to be a peer node to be demoted, and then the downgrade request message is sent to each peer node to be demoted.
  • the real-time status may also be a CPU idle ratio situation, and then the decision node first queries each peer node served by the client device.
  • the CPU idle ratio situation and then the peer node whose CPU idle ratio is higher than the predetermined value is determined as the peer node to be demoted, and then the downgrade request message is respectively sent to each of the peer nodes to be demoted.
  • Step S420 The peer node to be degraded, if it receives the leave consent message sent by the successor node, demotes itself to the client node, where the leave consent message is that the successor node can determine that the process can be processed according to its current processing capability. Sent after the service of the peer node is downgraded.
  • the successor node of the peer node to be demoted determines whether the service of the peer node to be demoted can be processed according to its current processing capability. If it is determined that the processing can be processed, the leaving consent message is sent to the to-be-degraded peer node, and if it is determined that the processing cannot be processed, the leaving rejection message may be sent to the to-be-degraded peer node.
  • the specific implementation process of determining whether the subsequent node can process the service of the peer node to be demoted according to its current processing capability will be described in detail in Embodiment 3 below.
  • the peer node to be demoted may downgrade itself to the client node after receiving the leave consent message sent by the successor node, and receive the leave reject message sent by the successor node. , the degradation process is not performed; if the peer node to be demoted has multiple successor nodes, the peer node to be demoted may downgrade itself to the client node after receiving the leave consent message sent by each successor node, that is, It is said that if the degraded peer node receives the leave reject message sent by one or more successor nodes, the downgrade process is not performed.
  • a peer node to be demoted has three successor nodes, namely, a successor node 1, a successor node 2, and a third node.
  • the messages sent by each successor node to the peer node to be demoted are shown in Table 1:
  • Point 1 and successor node 2 send the leave consent message, and a successor node (successor node 3) sends the leave reject message.
  • the peer node to be demoted may not perform the downgrade process.
  • the peer node to be demoted may not perform the downgrade process, end the downgrade process, or wait for a preset time length (for example, the preset time length is 10)
  • the leaving request message is sent back to the successor node of the peer node to be demoted, that is, step S410 and step S420 are repeatedly performed until the leaving consent message sent by the successor node is received.
  • the process for the degraded peer node to downgrade itself from the peer node to the client node may be, but is not limited to, the following:
  • the peer node to be demoted If the peer node to be demoted has only one successor node, the peer node to be demoted directly sends a storage request message to the successor node, and the storage request message carries the exchange resource stored by the peer node to be demoted, and the successor node stores The switching resource carried in the request message is stored, and the storage success message is sent to the to-be-degraded peer node. After the degraded peer node receives the storage success message sent by the successor node, it sends an update request to the predecessor node and the successor node respectively.
  • the degraded peer node sends an update completion message. After the degraded peer node receives the update completion message sent by each of the predecessor nodes and the successor node respectively, it is confirmed that the degraded process ends, and the to-be-degraded peer node is successfully demoted to the client node. After the degraded peer node is successfully downgraded to the client node, it can choose to exit the peer-to-peer network.
  • the peer node to be demoted first determines, for each subsequent node, according to a preset sharing rule, respectively, the subsequent node stored in the peer node to be demoted is responsible for And processing the exchange resource, and then sending a storage request message to the successor node, where the storage request message carries the exchange resource that is processed by the successor node, and each subsequent node stores the exchange resource carried in the request message, and waits for the
  • the degraded peer node sends a storage success message, and after the degraded peer node receives the storage success message sent by each subsequent node, sends an update request message to each of the predecessor node and the successor node, respectively, to notify each node to perform a routing table update operation, where If the to-be-degraded peer node receives the storage failure message sent by at least one successor node, it may be preset After the inter-length, the storage request message is sent to each subsequent node again.
  • each node After receiving the update request message, each node performs a routing table update operation, and then, after successfully updating the routing table, sends an update completion message to the to-be-degraded peer node, and the degraded peer node receives each of the predecessor nodes and the successor node respectively. After the update completion message is sent, it is confirmed that the downgrade process ends, and the to-be-degraded peer node is successfully downgraded to the client node.
  • the successor node of the peer node to be demoted is the successor node 1, the successor node 2, and the successor node 3.
  • the peer node to be demoted stores the exchange resource 1.
  • the exchange resource 2 the exchange resource 3, the exchange resource 4, the exchange resource 5 And exchanging resources.
  • the pre-defined sharing rules are as follows: If the to-be-degraded peer node stores N exchange resources, the successor node shares the first N/3 exchange resources, and the successor node 2 shares the intermediate N/3. Exchange resources, the successor node three share N/3 exchange resources.
  • the exchange resources shared by the successor nodes of the pair to be downgraded are shown in Table 2:
  • the peer to peer node exchanges the exchange resource 1 and the exchange resource 2 in the storage request message to the successor node 1 , and carries the exchange resource 3 and the exchange resource 4 in the storage request message and transmits the message to the successor node 2
  • the exchange resource 5 and the exchange resource 6 are carried in the storage request message and transmitted to the successor node 3.
  • the peer node to be demoted does not directly downgrade after receiving the leave consent message sent by the successor node, but sends a downgrade consent message to the decision node.
  • the degraded peer node receives the leave reject message sent by the successor node, and the peer node to be degraded sends a degraded reject message to the decision node, and the decision node of the peer network sends separately to each peer node to be demoted to send the downgrade consent message.
  • the degraded peer node starts to enter the demotion process after receiving the departure indication message sent by the decision node.
  • the degrading process of the peer node to be degraded in the case of passive degrading is the same as the degrading process of the peer node to be demoted in the case of active degrading, and is not mentioned here.
  • the peer node after the peer node is successfully demoted to the client node, the peer node may be selected. Exiting the peer-to-peer network, you can also choose to continue to exist in the peer-to-peer network as the client node.
  • the third embodiment of the present invention provides a peer-to-peer network node, where the peer-to-peer network node is a peer node to be demoted, and its structure is as shown in FIG. 9, which includes:
  • the leaving request message sending unit 31 is configured to send a leaving request message to a successor node of the peer network node;
  • the leave consent message receiving unit 32 is configured to receive the leave consent message sent by the successor node; the downgrade unit 33 is configured to: after the leave consent message receiving unit 32 receives the leave consent message, the peer network node is configured by The peer node is demoted to a client node, where the leave consent message is sent by the successor node after determining that the service of the peer network node can be processed according to its current processing capability.
  • the peer-to-peer network node further includes:
  • the downgrade request message receiving unit is configured to receive a downgrade request message sent by the decision node in the peer to peer network before the leave request message sending unit 31 sends the leave request message to the successor node of the peer network node.
  • the peer to peer network node further includes:
  • a first leaving consent message sending unit configured to send the leaving consent message to the decision node before the downgrading unit 33 demotes the peer network node from the peer node to the client node; the leaving indication message receiving unit, And the step-down unit 33 is further configured to demote the peer-to-peer network node from the peer node to the client node after the leave indication instruction receiving unit receives the leave indication message.
  • the peer-to-peer network node further includes:
  • the leaving rejection message receiving unit is configured to receive the leaving rejection message sent by the successor node; the leaving request message sending unit 31 is further configured to: after waiting for the preset time length after receiving the leaving rejection message, the leaving rejection message receiving unit And sending a leave request message to the successor node of the peer network node, where the leave reject message is sent by the successor node after determining that the service of the peer network node cannot be processed according to its current processing capability.
  • the peer node of the peer network node has multiple;
  • the degrading unit 33 is specifically configured to downgrade the peer network node from the peer node to the client node when the leaving consent message receiving unit 32 receives the leaving consent message respectively sent by each of the successor nodes of the peer network node. ;
  • the leaving request message sending unit 31 is specifically configured to: when the leaving reject message receiving unit receives the leaving reject message sent by the at least one successor node of the peer network node, wait for a preset time length, and then re-route to the peer-to-peer network The successor node of the node sends a leave request message.
  • Embodiment 4
  • FIG. 10 it is a schematic flowchart of a peer-to-peer network node degradation method proposed by the fourth embodiment of the present invention, and the specific processing procedure is as follows:
  • Step S510 After receiving the leave request message sent by the peer node to be demoted, the successor node of the peer node to be demoted in the peer-to-peer network determines whether the service of the peer node to be demoted can be processed according to its current processing capability. ;
  • the processing capabilities of the successor nodes may be, but are not limited to, characterized by the processing capabilities of the various processing resources, which may be, but are not limited to, include CPU resources, memory resources, and bandwidth resources.
  • the successor node can directly read the current processing capability of these processing resources, such as reading the current memory idle value and the bandwidth idle value, and for the CPU resources, since the successor node cannot directly read the current CPU processing. Capability, can only read the current CPU idle ratio, so the successor node needs to run the pre-downloaded benchmark test program to evaluate its overall CPU processing power before accessing the peer-to-peer network, and subsequently determine whether it can be processed.
  • the current CPU processing capability is determined according to the estimated CPU overall processing capability and the read current CPU idle ratio.
  • the successor node After the successor node completes the evaluation of the overall processing power of the CPU by running the pre-downloaded benchmark program, if the interaction information between the peer nodes is found, the version of the benchmark program of the other peer nodes is found to be higher than the version of the benchmark program. That is to say, the benchmark test program of the successor node needs to be updated, and the successor node can download the latest benchmark test program and evaluate the overall processing power of the CPU according to the re-downloaded benchmark test program. If the current load of the successor node is light, the overall CPU processing capability can be evaluated immediately after downloading the latest benchmark test program. If the current load of the successor node is heavy, the latest benchmark test program can be downloaded. Wait a while before evaluating the overall CPU processing power.
  • the successor node determines whether it is capable of processing the service of the peer node to be demoted, and not only needs to obtain the processing capability of each current processing resource, but also needs to acquire the processing capability required for processing each resource of the peer node to be downgraded, and then Each processing resource determines whether the processing capability of the current processing resource of the current processing device satisfies the processing capability required by the processing resource when processing the service of the peer node to be demoted, and if the determination result is yes, it is confirmed that the processing capability can be processed. Declaring the service of the peer node to be downgraded, otherwise, confirming that the service of the peer node to be demoted cannot be processed,
  • the process of determining, by the successor node, the processing capability required for each processing resource when processing the service to be demoted to the peer node may be, but is not limited to, the following: the successor node determines the waiting according to a preset sharing rule. Degrading the exchange resources stored in the peer node and being processed by the successor node, and then determining, according to the determined exchange resources, the processing capability required for each processing resource when processing the user request for the exchange resources, that is, processing the to-be-degraded pair The processing power required for each processing resource when the nodes are in business.
  • Step S520 If the result of the determination in step S510 is yes, the successor node sends a leave consent message to the to-be-degraded peer node, indicating that the to-be-degraded peer node is demoted to the client node.
  • the successor node may send a leave rejection message to the to-be-degraded peer node, indicating that the to-be-degraded peer node cannot be downgraded to the client node.
  • the peer node to be demoted in the peer-to-peer network first sends a leave request message to its successor node, and the successor node determines whether it can process the to-be-degraded pair according to its current processing capability.
  • the service of the peer node if it can be processed, sends a leave consent message to the peer node to be demoted, and the peer node degrades itself to the client node only after receiving the leave consent message, as seen from above, the peer-to-peer network
  • the peer node to be demoted is no longer to send the leave request message to the successor node and then directly degrades and then exits the network, but declassifies itself to the client node after the successor node agrees, that is, the peer node to be demoted only
  • Degraded processing will be performed when the successor node can process the service of the peer node to be demoted, thereby effectively avoiding the direct degraded peer node in the peer-to-peer network to the client node, causing the load of its successor node to be too heavy, or even causing The situation of peer-to-peer networks.
  • Embodiment 5 the fifth embodiment of the present invention provides a peer-to-peer network node, where the peer-to-peer network node is a successor node of the peer node to be demoted, and its structure is as shown in FIG.
  • the leave request message receiving unit 51 is configured to receive the leave request message sent by the peer node to be demoted
  • the determining unit 52 is configured to determine, according to the current processing capability of the peer network node, whether the service of the peer node to be demoted can be processed;
  • the second leaving consent message sending unit 53 is configured to send a leaving consent message to the to-be-degraded peer node when the determining result of the determining unit 52 is YES, indicating that the to-be-degraded peer node is demoted to a client node.
  • the determining unit 52 specifically includes:
  • a processing capability determining subunit configured to determine a processing capability of each of the processing resources of the peer network node and a processing capability required by each processing resource when processing the service of the peer node to be demoted; Determining, by each processing resource, whether the processing capability of the current processing resource of the peer network node meets the processing capability required by the processing resource when processing the service of the peer node to be demoted;
  • a confirmation subunit configured to: when the judgment result of the judgment subunit is yes, confirm that the peer network node can process the service of the peer node to be demoted, and when the judgment result of the subunit is at least And confirming that the peer network node is unable to process the service of the peer node to be demoted.
  • the method for implementing the sixth embodiment of the present invention is as follows: In the case that the peer node to be demoted has only one successor node, the method for degrading the peer network node in the active demotion is implemented in a flowchart. The specific processing flow is as follows:
  • Step S610 The peer node to be demoted in the peer-to-peer network sends a leave request message to its successor node.
  • Step S620 The successor node determines, according to its current processing capability, whether the service of the peer node to be demoted can be processed.
  • Step S630 if the successor node determines that the service of the peer node to be demoted can be processed, The to-be-degraded peer node sends a leave consent message, and if it is determined that the service of the peer node to be demoted cannot be processed, sending a leave rejection message to the to-be-degraded peer node;
  • Step S640 if the degraded peer node receives the leave consent message sent by the successor node, the process of the downgrade process is performed, and the save request message is sent to the successor node, where the store request message carries the node to be demoted to be stored.
  • the to-be-degraded peer node may wait for the preset time length to send the leave request message to the successor node again.
  • Step S650 the successor node stores the exchange resource carried in the request message.
  • Step S660 after the successor node successfully stores the exchange resource, sends a storage success message to the peer node to be demoted, and if the subsequent node fails to store, the node waits Degrading the peer node to send a storage failure message;
  • Step S670 if the degraded peer node receives the storage success message sent by the successor node, sending an update request message to each of the predecessor node and the successor node respectively, informing each node to perform a routing table update operation ;
  • the to-be-degraded peer node may wait for the preset length of time to send a storage request message to the successor node again.
  • Step S680 The predecessor node and the successor node of the peer node to be demoted perform the routing table update operation after receiving the update request message.
  • Step S690 The predecessor node and the successor node of the peer node to be degraded send an update completion message to the to-be-degraded peer node after successfully updating the routing table.
  • the method for degrading the peer network node in the passive degradation is implemented in a flowchart.
  • the specific processing flow is as follows:
  • Step S710 The decision node in the peer-to-peer network performs real-time status query on each peer node served by the client device, and determines each peer node to be demoted according to the real-time status of the query;
  • Step S720 The decision node sends a demotion request message to each of the determined peer nodes to be degraded respectively.
  • Step S730 the peer node to be demoted to the downgrade request message is received, and the peer node is to be demoted.
  • the successor node of the point sends a leave request message;
  • Step S740 the successor node determines, according to its current processing capability, whether the service of the peer node to be demoted can be processed
  • Step S750 if the successor node determines that the service of the peer node to be demoted can be processed, sending a leave consent message to the peer node to be demoted, and if it is determined that the service of the peer node to be demoted cannot be processed, The degraded peer node sends a leave reject message;
  • Step S760 if the degraded peer node receives the leave consent message, sends a downgrade consent message to the decision node, and if the leave reject message is received, sends a downgrade reject message to the decision node; step S770, the decision node sends a downgrade consent message.
  • Step S780 after the degraded peer node receives the leaving indication message sent by the decision node, the process of the degrading process is performed, where the process of the degrading process is consistent with the process of the degrading process of step S640 to step S690 in the sixth embodiment of the present invention, and details are not described herein again. .
  • Example eight
  • Embodiment 8 of the present invention provides a peer-to-peer network node degradation system, which has the structure shown in FIG. 14, and includes a peer node 82 to be demoted, and a successor node 82 of the peer node 81 to be demoted, wherein:
  • the degraded peer node 81 is configured to send a leave request message to its successor node 82, and if it receives the leave consent message sent by the successor node 82, demote itself to the client node;
  • the successor node is configured to determine, according to its current processing capability, whether the service of the peer node 8 1 to be degraded can be processed after receiving the leave request message sent by the peer node 81 to be demoted, if the judgment result is yes And sending a leave consent message to the to-be-degraded peer node 81.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer And Data Communications (AREA)

Abstract

L'invention concerne un procédé et un système d'accès à des nœuds P2P. Le procédé comprend les étapes suivantes : acquisition par un nœud homologue en surcharge ou un serveur de gestion d'une valeur de performances en temps réel d'un nœud client candidat en liaison directe avec le nœud homologue en surcharge; détermination d'un nœud client candidat caractérisé en ce que la valeur de performances en temps réel est supérieure à une valeur seuil prédéfinie de performances en tant que nœud client à promouvoir; et sélection d'au moins un nœud client à promouvoir et promotion de celui-ci pour en faire un nœud homologue, de façon à résoudre le problème de l'échec de séparation de charges. L'invention concerne de plus un procédé et un système de rétrogradation de nœuds de réseau homologue, ainsi qu'un nœud de réseau homologue. Le procédé comprend les étapes suivantes : un nœud homologue à rétrograder envoie un message de demande de détachement à un nœud ultérieur de celui-ci; si un message d'approbation de détachement envoyé par le nœud ultérieur est reçu, alors le nœud homologue à rétrograder se rétrograde en nœud client, le message d'approbation de détachement étant envoyé par le nœud ultérieur lorsqu'il estime pouvoir traiter le service du nœud homologue à rétrograder en fonction de sa capacité actuelle de traitement, de façon à résoudre le problème selon lequel un nœud homologue se rétrogradant directement en nœud client provoque une surcharge de son nœud ultérieur.
PCT/CN2012/082415 2011-10-28 2012-09-29 Procédé d'addition de nœuds pour système p2p, système p2p correspondant, procédé et système de rétrogradation de nœuds d'un réseau homologue, et nœud de réseau homologue WO2013060221A1 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201110334787.1A CN103095744B (zh) 2011-10-28 2011-10-28 对等网络节点降级方法及系统、对等网络节点
CN201110333816.2 2011-10-28
CN201110333816.2A CN103095742B (zh) 2011-10-28 2011-10-28 用于p2p系统的节点加入方法及相应的p2p系统
CN201110334787.1 2011-10-28

Publications (1)

Publication Number Publication Date
WO2013060221A1 true WO2013060221A1 (fr) 2013-05-02

Family

ID=48167112

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/082415 WO2013060221A1 (fr) 2011-10-28 2012-09-29 Procédé d'addition de nœuds pour système p2p, système p2p correspondant, procédé et système de rétrogradation de nœuds d'un réseau homologue, et nœud de réseau homologue

Country Status (1)

Country Link
WO (1) WO2013060221A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101064649A (zh) * 2007-02-02 2007-10-31 华为技术有限公司 选举超级节点、搜索网络节点或资源的方法、装置及系统
CN101521679A (zh) * 2009-04-03 2009-09-02 南京邮电大学 基于混合式结构化对等网络的自组织方法
CN101599995A (zh) * 2009-07-13 2009-12-09 中国传媒大学 面向高并发检索系统的目录分发方法及网络架构
CN101729277A (zh) * 2008-10-27 2010-06-09 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 (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101064649A (zh) * 2007-02-02 2007-10-31 华为技术有限公司 选举超级节点、搜索网络节点或资源的方法、装置及系统
CN101729277A (zh) * 2008-10-27 2010-06-09 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
CN101521679A (zh) * 2009-04-03 2009-09-02 南京邮电大学 基于混合式结构化对等网络的自组织方法
CN101599995A (zh) * 2009-07-13 2009-12-09 中国传媒大学 面向高并发检索系统的目录分发方法及网络架构

Similar Documents

Publication Publication Date Title
CN106656800B (zh) 一种路径选取方法及系统、网络加速节点及网络加速系统
US10033548B2 (en) Method, system, service selection entity, and service management entity for selecting service provision entity
US9075660B2 (en) Apparatus and method for providing service availability to a user via selection of data centers for the user
CN111464649B (zh) 一种访问请求回源方法和装置
KR20090069312A (ko) 리소스 공급을 위한 방법, 시스템, 및 에지 서버
US20090323700A1 (en) Method of determining a routing path
CN102497409B (zh) 一种云计算系统资源管理的方法
US8156217B2 (en) Dynamically balancing load for servers
Varghese et al. Edge-as-a-service: Towards distributed cloud architectures
JP5011433B2 (ja) 他のペアの近傍にあるペアグループを決定する方法、関連するサーバ及び解析装置
JP2020087443A (ja) 企業環境のための動的コンテンツ配信プロトコル
JP2010537522A (ja) 通信ネットワークシステム及び通信ネットワークのサービス処理方法
Park et al. Smart base station-assisted partial-flow device-to-device offloading system for video streaming services
CN114257594B (zh) 分布式系统中对用户网络侧进行网络资源分配的方法
CN113783741B (zh) 用于配置网关设备的方法、装置、服务器、网关设备及系统
KR101883671B1 (ko) 노드 분산 방법 및 이를 수행하는 관리 서버
Saifullah et al. Open flow-based server load balancing using improved server health reports
WO2013060221A1 (fr) Procédé d'addition de nœuds pour système p2p, système p2p correspondant, procédé et système de rétrogradation de nœuds d'un réseau homologue, et nœud de réseau homologue
JP2019041241A (ja) 振り分けシステム
CN114900526A (zh) 负载均衡方法及系统、计算机存储介质、电子设备
JP2012169789A (ja) 負荷分散サーバ及びサーバ選択方法及びサーバ選択プログラム
US20210168074A1 (en) Methods and systems for identifying optimal network paths within an association
JP2016046785A (ja) キャッシュサーバ選択装置、分散キャッシュシステム、及びキャッシュサーバ選択方法
JP2013538383A (ja) 複数基準の選択を有するトポロジサーバを用いて通信アーキテクチャにわたって分散されたノードのネットワークへのアクセス
JP2011048504A (ja) 認証サーバ間の負荷分散方法及び装置及びプログラム

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: 12843422

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: 12843422

Country of ref document: EP

Kind code of ref document: A1