WO2010115373A1 - 基于对等网络的资源信息备份操作方法及对等网络 - Google Patents

基于对等网络的资源信息备份操作方法及对等网络 Download PDF

Info

Publication number
WO2010115373A1
WO2010115373A1 PCT/CN2010/071617 CN2010071617W WO2010115373A1 WO 2010115373 A1 WO2010115373 A1 WO 2010115373A1 CN 2010071617 W CN2010071617 W CN 2010071617W WO 2010115373 A1 WO2010115373 A1 WO 2010115373A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
domain
backup
resource
information
Prior art date
Application number
PCT/CN2010/071617
Other languages
English (en)
French (fr)
Inventor
彭永林
陶全军
张永辉
裘晓峰
雷震宇
张春红
李漓春
王岩
弭伟
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 WO2010115373A1 publication Critical patent/WO2010115373A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • H04L67/104Peer-to-peer [P2P] networks

Definitions

  • the present invention relates to a peer-to-peer network, and more particularly to a resource information backup operation method and a peer-to-peer network based on a peer-to-peer network.
  • Peer to Peer (P2P) network as the overlay network of the existing Internet (Internet), shares computer resources and services through peer-to-peer exchange between systems to achieve distributed, reliable, scalable, and Robust web application model. It breaks the traditional client/server (C/S) model and the status of each node in the network is equal. Each node acts as a server, serving other nodes, and also enjoying the services provided by other nodes.
  • C/S client/server
  • DHT Distributed Hash Table
  • Overlay Network P2P networks
  • some nodes may leave the network normally or abnormally, which may easily cause loss of storage resources. In order to enhance the availability of storage resources, they may not be obtained due to the failure of individual nodes, usually in a DHT network. Use other nodes to make some backups for these resources.
  • the node used to back up a resource is called the backup node of this resource. Backup resources increase the availability of resources.
  • the backup of resources is mainly performed inside a DHT network.
  • a node forms a DHT ring according to its node ID from small to large clockwise, taking the first node that a node encounters in a clockwise direction as its backward node, since each node maintains the node.
  • the data stored by the node is directly backed up on its backward node. If multiple copies need to be stored, it is stored on multiple nodes behind it.
  • This kind of backup method is convenient to maintain, but in some abnormal cases, this method does not guarantee the availability of resources.
  • the internal backup mode of the DHT network has been improved, mainly for two aspects: First, by copying multiple copies of data and storing them on several nodes according to certain rules, only one data can be retrieved during recovery; The data is divided into multiple redundant segments, so that only a few segments can be retrieved to recover the data, so that the backup of the data is relatively small, but the maintenance overhead is relatively large.
  • the technical problem to be solved by the present invention is to provide a resource information backup operation method and a peer-to-peer network based on a peer-to-peer network to better ensure data reliability and availability.
  • the present invention provides a resource information backup operation method based on a peer-to-peer network, the node of the peer-to-peer network having a node global identifier including a domain identifier and a node identifier, and resource information of the peer-to-peer network
  • the method includes: a resource global identifier including a domain identifier and a resource identifier, where the resource information is saved on a primary node that has the same domain identifier as the resource information, and the method for performing the backup operation on the resource information includes:
  • the initiating node sends an out-of-domain backup node determining request to the connection node of the backup domain, where the out-of-domain backup node determines the resource global identifier information including the resource information of the backup operation, where the connection node of the backup domain and the primary node have Different domain identifiers;
  • the connection node of the backup domain determines the out-of-domain backup node according to the resource global identifier information and the out-of-domain backup rule, and sends the routing information of the out-of-domain backup node to the initiating node; the initiating node according to the routing information Sending an out-of-domain backup operation request to the out-of-domain backup node, and the out-of-domain backup node performs corresponding processing according to the out-of-domain backup operation request.
  • the method further includes: if the out-of-domain backup operation request is an out-of-domain backup request, the initiating node is a primary node where the resource information is located, and the domain node sends an out-of-domain backup request to the out-of-domain backup node.
  • Resource information content carrying the resource information, the extra-domain backup section The point backs up the resource information content and the resource global identifier information according to the extra-domain backup request;
  • the originating node is a query node, and the extra-domain backup node returns the resource to be queried according to the resource global identifier information in the extra-domain backup query request to the query node.
  • Resource information content of the information
  • the initiating node is a node that detects that the primary node of the resource information is faulty, and the resource global identifier information refers to a resource global identifier of the resource information to be restored. Or a resource global identifier range, the resource global identifier or the resource global identifier range is determined by the initiating node according to the intra-domain resource storage rule, and the extra-domain backup node sends the initiating node according to the extra-domain backup recovery request Returning the resource information content corresponding to the resource global identifier or the resource information content corresponding to the resource global identifier in the range of the resource global identifier, and saving by the initiating node.
  • the method further includes: the node of the peer-to-peer network storing a resource information table, an intra-domain routing table, and an out-of-domain routing table, where the resource information table includes a resource global identifier of the resource information and resource information content, and the intra-domain routing table a node global identifier and routing information of an intra-domain node having a connection relationship with the node, where the extra-domain routing table includes a node global identifier and routing information of an extra-domain node having a connection relationship with the node, and the extra-domain backup node It is determined by combining the global identifier of the node in the routing table in the domain.
  • the present invention further provides a peer-to-peer network
  • the node of the peer-to-peer network has a node global identifier including a domain identifier and a node identifier
  • the resource information of the peer-to-peer network includes a domain identifier and a resource.
  • the identified resource global identifier, the node of the peer-to-peer network includes a receiving module, an out-of-domain backup node determining module connected to the receiving module, and a processing module connected to the receiving module and the extra-domain backup node determining module, where:
  • a receiving module configured to receive an out-of-domain backup node determining request sent by the initiating node, where the out-of-domain backup node determines resource global identifier information including resource information to be backed up in the request, and the connecting node of the backup domain and the primary node Having different domain identifiers; also for accepting an out-of-domain backup operation request sent by the originating node;
  • An out-of-domain backup node determining module configured to determine an out-of-domain backup node according to the resource global identifier information and an out-of-domain backup rule;
  • the processing module is configured to send the routing information of the out-of-domain backup node to the initiating node according to the determined out-of-domain backup node; and is further configured to perform corresponding processing according to the out-of-domain backup operation request.
  • the initiating node is a primary node where the resource information is located, and the out-of-domain backup operation request further carries resource information content of the resource information, where
  • the processing module is configured to back up the resource information content and the resource global identifier information according to the extra-domain backup operation request;
  • the initiating node is a query node
  • the processing module is configured to return, to the query node, the resource information to be queried according to the resource global identifier information. information;
  • the initiating node is a node that detects that the primary node of the resource information is faulty, and the global identifier information of the resource in the out-of-domain backup recovery request is to be restored.
  • a resource global identifier or a resource global identifier range of the resource information where the resource global identifier or the resource global identifier range is determined by the initiating node according to the intra-domain resource storage rule, and the processing module is configured to use
  • the out-of-domain backup recovery request returns the resource information content corresponding to the resource global identifier or the resource information content corresponding to the resource global identifier in the range of the resource global identifier to the originating node.
  • the node further includes a storage module that is connected to the extra-domain backup node determining module and the processing module, and is configured to store a resource information table, an intra-domain routing table, and an extra-domain routing table, where the resource information table includes resource resources globally.
  • the global identifier and the routing information are determined by combining the global identifier of the node in the routing table in the domain.
  • the present invention further provides a resource information backup operation method based on a peer-to-peer network, where a node of the peer-to-peer network has a node global identifier including a domain identifier and a node identifier, and resources of the peer-to-peer network
  • the information has a resource global identifier including a domain identifier and a resource identifier, and the resource information is saved on a master node having the same domain identifier as the resource information
  • the method for performing the backup operation on the resource information includes:
  • the initiating node sends an out-of-domain backup operation request to the connection node of the backup domain, where the out-of-domain backup node operation request includes a resource global identifier of the resource information to be backed up, and the backup domain
  • the connection node has a different domain identifier than the master node
  • connection node of the backup domain determines the out-of-domain backup node according to the resource global identifier and the out-of-domain backup rule, and forwards the out-of-domain backup operation request to the out-of-domain backup node, where the routing information of the initiating node is carried;
  • the out-of-domain backup node performs corresponding processing according to the out-of-domain backup operation request.
  • the method further includes: the out-of-domain backup operation request is an out-of-domain backup request, a backup query request, or an out-of-domain backup recovery request; if the out-of-domain backup operation request is an out-of-domain backup request, the request further carries the resource information content to be backed up.
  • the present invention further provides a peer-to-peer network, the node of the peer-to-peer network having a node global identifier including a domain identifier and a node identifier, and storing a node global identifier of other nodes in the domain,
  • the resource information of the peer-to-peer network has a resource global identifier including a domain identifier and a resource identifier, where the resource information is saved on a node having the same domain identifier, and is backed up on an out-of-domain backup node having a different domain identifier.
  • the extra-domain backup node is determined according to the resource global identifier of the resource information, the node global identifier of the backup domain, and the extra-domain backup rule.
  • the above method further includes: the resource information is simultaneously backed up on other nodes having the same domain identifier.
  • the method and the peer-to-peer network of the present invention perform extra-domain backup of resource information to better ensure data reliability and availability; and additionally perform extra-domain backup for resource information, so that there are nodes in the local network.
  • FIG. 1 is a schematic diagram of a P2P network architecture of the present invention
  • FIG. 2 is a schematic diagram of global identifiers of node and resource information in the present invention.
  • FIG. 3 is a schematic diagram of content stored by a node in the present invention.
  • FIG. 4 is a schematic flowchart of a resource information backup operation method according to the present invention.
  • 5 is a schematic diagram of a backup process of resource information in the application example 1 of the method of the present invention;
  • FIG. 6 is a schematic diagram of searching for backup resource information in the application example 2 of the method of the present invention;
  • FIG. 7 is a schematic diagram of a method for applying a new node to a method of applying the method of the present invention.
  • FIG. 9 is a schematic structural diagram of a module of a peer to peer network node according to the present invention.
  • the resource backup method of the present invention is implemented based on a geographically optimized P2P network architecture.
  • the network architecture is briefly introduced below:
  • the entire P2P network is divided into several areas according to the geographic location, and the area identifier of each area (ie, the domain identifier, referred to as the domain ID) is determined.
  • the nodes with the same domain identifier form a domain, and the nodes in the domain follow a certain DHT.
  • the algorithm forms a sub-P2P network, and each node in the area keeps in contact with other domains according to certain rules (see Figure 1).
  • Each node (or resource) in the network has a global node-only node (or resource) global identifier.
  • this global identifier consists of a domain identifier and a node (or resource) identifier.
  • the node and resource are based on Its local domain determines its domain identity, and the node (or resource) identity is obtained by hashing some information about the node (or resource) such as node IP, port number (or resource name).
  • the insertion, lookup, etc. of nodes (or resources) in the network are performed based on the global identifier of the node (or resource).
  • each resource is stored on the node of the domain (that is, the domain with the same domain as itself), and the global identifier of this node should be equal or similar to the resource global identifier of the resource (generally taken later) To the nearest).
  • the node identifier is obtained by hashing the feature information of the node, and the feature information of the node refers to the unique information of the node, such as an IP address, a network card number, or a machine serial number, and the selection of the hash algorithm may be arbitrary;
  • the resource identifier is obtained by performing a hash algorithm on the feature information of the resource information, and the feature information of the resource is information that can identify the resource, such as the specific content of the resource, the name of the user, the name of the user data, and the like, and the hash algorithm Selection can be arbitrary.
  • the content that needs to be stored in each node is as shown in FIG. 3, and includes a routing information table, a resource information table, and a domain ID mapping table.
  • the routing information table includes an intra-domain routing table of the domain in which the node is located and an extra-domain routing table of the extra-domain node routing information that is in contact with the node;
  • the resource information table includes resource information stored by the node and backup resource information stored therein.
  • the backup resource information mentioned here includes backup of extra-domain resource information and backup of intra-domain resource information.
  • the resource information stored in the resource information table includes a resource global identifier of the resource information and resource information content.
  • the present invention not only performs data backup in the local domain but also outside the domain when backing up resources. (The domain ID is different from the domain ID of the resource.)
  • the backup is performed to meet the requirements of remote disaster recovery while maintaining the advantages of intra-domain backup, and data recoverability is guaranteed.
  • Master node A node whose resource information is saved in the domain according to a storage rule; specifically, the storage rule is: a resource is stored on a node, a node ID of the node, and the resource The resource ID of the information is the closest and greater than or equal to the resource ID. If there is no node ID larger than the resource ID, it is stored on the node with the smallest node ID in the domain.
  • This domain the domain where the primary node is located
  • the local domain backup node Backs up the local node information of the primary node resource information according to a certain local domain backup rule.
  • the local domain backup rule may be: The backward node of the primary node serves as the local domain backup node.
  • the nodes are arranged in a certain order according to the size of the node ID (generally in ascending order), which is called a DHT ring (as shown in FIG. 5), and takes the first node that a node encounters in a clockwise direction as a node.
  • the backward node in the counterclockwise direction, encounters the first node as its forward node, and each node has a forward node and a backward node.
  • Backup domain The domain used for backup of the domain data.
  • the backup domain selection rule is generally selected as the backup domain with the domain size or the domain with a relatively low load.
  • the backup domain can be selected by manually specifying the mapping between the domain and the backup domain. Tables, you can also use some algorithms to select the backup domain of this domain according to the size of the domain.
  • Out-of-domain backup node An out-of-domain node that backs up resource information according to certain off-domain backup rules. Choose When the extra-domain backup node is selected, the resource may be backed up to the corresponding backup node of the backup domain according to the resource ID of the domain resource information. For example, the extra-domain backup node may be equal or closest to the resource ID in the backup domain. The node, which is greater than or equal to the first node of the resource ID; when the data is added, deleted, or updated, the backup data of the backup domain is changed accordingly.
  • the invention backs up the resource information according to the resource identifier.
  • the domain node changes, the corresponding backup data stored on the foreign domain node does not need to be adjusted. Only changes in resource information (such as resource information added, deleted, or changed) require corresponding changes in the foreign domain.
  • the method for the resource information backup operation of the peer-to-peer network includes: Step 401: The originating node sends an extra-domain backup node to the connection node of the backup domain to determine a request; and the extra-domain backup node determines that the request includes the backup. Resource global identifier information of the resource information of the operation, the connection node of the backup domain and the master node have different domain identifiers;
  • the resource global identifier information referred to herein may be a determined resource global identifier or a resource global identifier range.
  • Step 402 The connection node of the backup domain determines the extra-domain backup node according to the resource global identifier information and the extra-domain backup rule, and sends the routing information of the extra-domain backup node to the initiating node.
  • Step 403 The initiating node sends an out-of-domain backup operation request to the out-of-domain backup node according to the routing information, and the out-of-domain backup node performs corresponding processing according to the out-of-domain backup operation request.
  • the originating node is a primary node where the resource information is located, and the extra-domain backup request sent by the primary node to the extra-domain backup node further carries the resource information.
  • the resource information content, the extra-domain backup node backs up the resource information content and the resource global identifier information according to the extra-domain backup request;
  • the originating node is a query node, and the extra-domain backup node returns the resource to be queried according to the resource global identifier information in the extra-domain backup query request to the query node.
  • Resource information content of the information
  • the initiating node is a node that detects that the primary node of the resource information is faulty, and the resource global identifier information refers to recovery. a resource global identifier or a resource global identifier range of the resource information, where the resource global identifier or the resource global identifier range is determined by the initiating node according to the intra-domain resource storage rule, and the extra-domain backup node is configured according to the domain
  • the backup recovery request returns the resource information content corresponding to the resource global identifier or the resource information content corresponding to the resource global identifier in the range of the resource global identifier to the initiating node, and is saved by the initiating node.
  • the initiating node sends an out-of-domain backup operation request to the extra-domain backup node according to the routing information returned by the backup domain connection node, and of course, the backup domain connection node may forward the extra-domain backup operation request to the extra-domain backup node, where Initiating the routing information of the node, so that the processing by the out-of-band backup node returns the processing result to the initiating node.
  • the second method specifically includes:
  • Step A The initiating node sends an out-of-domain backup operation request to the connection node of the backup domain, where the out-of-domain backup node operation request includes resource global identifier information of the resource information to be backed up, and the connection node of the backup domain and the main Nodes have different domain identifiers;
  • Step B The connection node of the backup domain determines the extra-domain backup node according to the resource global identifier information and the extra-domain backup rule, and forwards the extra-domain backup operation request to the extra-domain backup node, where the resource information to be operated is carried.
  • Step C The out-of-domain backup node performs corresponding processing according to the out-of-domain backup operation request, and returns a processing result to the initiating node according to the routing information.
  • the out-of-domain backup operation request described above is an out-of-domain backup request, a backup query request, or an out-of-domain backup recovery request. If the out-of-domain backup operation request is an out-of-domain backup request, the request also needs to carry the resource information content to be backed up.
  • the primary node of the domain (the node that stores the original data of the resource) and the backup node of the domain (the node used for data backup in this domain) are all unavailable due to some abnormality, the data can be searched through the backup domain. And recovery.
  • the data search is performed according to the global identifier of the resource information.
  • the backup domain is searched for, and the request is a request for querying the backup data, and the request is Direct data query requests are queried differently;
  • a node of the domain detects this situation, and actively reports to the backup domain.
  • a data recovery request is issued, which includes a range of global resource identifiers for resource information that needs to be recovered.
  • the domain data is backed up by some intra-domain backup method.
  • the backup method in the domain can use the backup method of a single DHT network.
  • the selection of the backup method can be arbitrary.
  • Each node in Figure 5, 6, 7, and 8 is a home subscriber server (Home Subscriber Server,
  • HSS HSS is the primary user database that supports IMS network entities for handling calls/sessions. It contains user profiles, performs user authentication and authorization, and provides information about the user's physical location.
  • each user's user data called resource information, is stored in each HSS (hereinafter referred to as a node).
  • domain A For the sake of simplicity, the figure shows two fields: domain A and domain B, where nodes N7 ⁇ N46 belong to domain A, node n3 ⁇ ! 153 belongs to domain B, and the node number is its node ID.
  • Domain B is the backup domain of domain A. In the actual network, the backup domain is selected according to the size of the domain or the load status of the domain. When the backup domain is selected, the domain ID of the backup domain is specified when the new node joins, and the node of the A domain saves the information. Domain A also serves as a backup domain for other domains. For simplicity, the other fields are not shown in the figure.
  • Each node in the figure uses the One-hop DHT algorithm in the domain (that is, each node in the domain maintains a global routing table in the domain to maintain full connectivity with other nodes in the domain), and each node also maintains an out-of-domain Routing table, this out-of-domain routing table contains routing information for some nodes outside the domain.
  • the present invention is mainly for extra-domain backup.
  • the intra-domain backup the simplest form is used, that is, the backward node of the primary node is selected as the backup node, and the backup node is used for backing up all the resource information on the primary node (when there is new data) Keep up-to-date with backup data when joining or data updates).
  • the extra-domain backup rule in this embodiment is:
  • the out-of-domain node of the backup resource information has such a section A node identified by a point whose node identifier is equal to or greater than a resource identifier of the resource information and is closest to the resource identifier of the resource information.
  • Step 502 Node N7 sends a backup request to its backward node (node N12) to back up the new data;
  • Step 503 Node N12 backs up the new data and returns a response.
  • Step 504 The node N7 performs an out-of-domain backup on the newly added resource information, and sends an out-of-domain backup request to the connection node (for example, the node n53) corresponding to the backup domain, where the request includes a resource global identifier of the resource information to be backed up;
  • connection nodes of the backup domain are randomly selected.
  • Step 506 The node N7 issues an out-of-domain backup request to the node n9, where the request includes the resource global identifier of the resource information and the resource information content;
  • Step 507 The node n9 backs up the resource information according to the received request, and returns a backup success response.
  • Data search is divided into intra-domain data search and extra-domain backup data search.
  • the lookup process of extra-domain backup data is given.
  • the intra-domain data search the global identifier of the query resource information can be directly queried in the local domain according to the data storage rule. (For the lookup of the extra-domain backup data, it will be performed only when both the local domain data and the backup data of the local domain are unavailable.)
  • the query process includes:
  • Step 602 The node N23 sends a query for the backup data request to the backup node (node N12) in the domain because the node N7 is abnormally down or the request fails.
  • Step 603 The node N12 also generates some abnormality, the query request fails, and the node N23 issues an out-of-domain backup query request to a certain node (for example, the node n32) of the backup domain, where the request includes the global identifier of the resource to be searched.
  • a certain node for example, the node n32
  • the request for querying backup data here is different from the request for direct data query.
  • the global identifier of the resource is included in the request because the backup domain may still have backup data of other domains);
  • Step 604 The node n32 determines the extra-domain backup node of the resource according to the resource global identifier and the extra-domain backup rule, and searches for the intra-domain routing table to obtain the routing information of the extra-domain backup node (here, the node n9), and forwards the information to the node n9.
  • the outbound backup query request, the forwarded query request should carry the routing information of N23;
  • Step 605 The node n9 returns the resource information of the required query to the node N23.
  • the out-of-domain backup data lookup only occurs when the data master node and the backup node in the domain fail to query. This situation is usually caused by a large number of nodes in the domain being abnormally down, so that the data can be obtained through the lookup of the backup data outside the domain. Get it normally.
  • the data to be copied from the backward node includes three parts, as shown in Figure 7, the specific copying steps are as follows:
  • Step 701 The new node is responsible for saving the resource information (the resource ID value is the part of the resource information between the forward node ID and the new node ID), where the resource ID is (16, 20) (here indicated The node is the data range that the master node should be responsible for, and the node ID of the data should be greater than 16 and less than or equal to 20;
  • Step 702 The new node is responsible for saving the backup data as the backup node outside the domain (the resource ID saved in the backward node is less than or equal to the backup data of the new node ID of the domain). It is assumed here that the domain A is the backup domain of the domain C (not shown), and the backup resource information is the data of the resource ID in the domain C at (16, 20); Step 703: The new node is the backup data of the forward node that the backup node in the domain should be responsible for (the domain backup data saved in the backward node), where the resource ID is in the data of (12, 16).
  • the backward node N16 of the backup node in the domain detects that two or more consecutive nodes have failed (the node N16 detects the failure of its forward continuous multiple nodes through the keep-alive mechanism between it and its forward node), and the backup is based on the external domain.
  • the rule copies the resource information from the backup node outside the lost data domain and backs up the backup.
  • Step 801 The backward node N16 of the backup node in the domain detects the fault (assuming that only N7 and N12 fail), and only knows that the ID range of the lost resource information in the domain is [the forward node ID of the failed master node, and the failed master node ID], the failed master node is here node N7, and the lost resource information is resource information whose resource ID is located at (46, 7), that is, resource information whose resource ID is greater than 46 or not greater than 7;
  • the out-of-domain backup rule in this embodiment is:
  • the out-of-domain node of the backup resource information is a node having such a node identifier, and the node identifier is equal to or larger than the resource identifier of the resource information, and is closest to the resource identifier of the resource information.
  • the data recovery and backup that the node N16 needs to perform includes:
  • the resource information lost in the original domain is:
  • Step 803 The node N16 copies the required data from the outside of the domain, and the node N16 becomes the master node of the resource information;
  • Step 804 The node N16 re-domains the intra-domain backup data stored on the N16 by the data that has just been restored and the failed forward node of the node N16, that is, backs up the restored data and the data backed up in the domain to the node N23.
  • Step 805 The node N16 backs up the data stored by the node N46.
  • Step 806 The range of the lost out-of-domain backup data is: (forward node ID of the failed master node, node N16 forward node ID], that is, the out-of-domain backup data whose resource ID is in the range of (46, 12). If these backup data When multiple domains are involved, the node N16 needs to re-backup the required data to multiple domains. The specific process is the same as the data recovery of the failed node, and will not be described.
  • the present invention further provides a network architecture, and performs functional module division on the node.
  • the node of the peer-to-peer network includes a receiving module and an external backup connected to the receiving module. a node determining module, a processing module connected to the receiving module and the extra-domain backup node determining module, and the node further includes a storage module connected to the extra-domain backup node determining module and the processing module, where:
  • a receiving module configured to receive an out-of-domain backup node determining request sent by the initiating node, where the out-of-domain backup node determines resource global identifier information including resource information to be backed up in the request, and the connecting node of the backup domain and the primary node Having different domain identifiers; also for accepting an out-of-domain backup operation request sent by the originating node;
  • An out-of-domain backup node determining module configured to determine an out-of-domain backup node according to the resource global identifier information and an out-of-domain backup rule
  • the processing module is configured to send routing information of the out-of-domain backup node to the initiating node according to the determined out-of-domain backup node; and is further configured to perform corresponding processing according to the out-of-domain backup operation request.
  • the initiating node is the primary node where the resource information is located, and the extra-domain backup operation request further carries the resource information content of the resource information, and the processing module And configured to back up the resource information content and the resource global identifier information according to the out-of-domain backup operation request;
  • the initiating node is a query node
  • the processing module is configured to return, to the query node, the resource information to be queried according to the resource global identifier information. information;
  • the initiating node is a node that detects that the primary node of the resource information is faulty, and the global identifier information of the resource in the out-of-domain backup recovery request is to be restored.
  • a resource global identifier or a resource global identifier range of the resource information where the resource global identifier or the resource global identifier range is determined by the initiating node according to the intra-domain resource storage rule, and the processing module is configured to use
  • the out-of-domain backup recovery request returns the resource information content corresponding to the resource global identifier or the resource information content corresponding to the resource global identifier in the range of the resource global identifier to the originating node.
  • a storage module configured to store a resource information table, an intra-domain routing table, and an out-of-domain routing table, where the resource information table includes a resource global identifier of the resource information and a resource information content, where the intra-domain routing table includes a connection relationship with the node a node global identifier and routing information of the intra-domain node, the extra-domain routing table includes a node global identifier and routing information of the extra-domain node having a connection relationship with the node, and the extra-domain backup node is a node in the intra-domain routing table The global identifier is determined.
  • the data backup in the domain can ensure the recoverability of data when a small number of nodes in the network are normal or abnormally leave the network.
  • the present invention achieves the advantages of intra-domain backup and also realizes remote disaster recovery. Better guarantee the reliability and availability of data.
  • the present invention performs an extra-domain backup for resource information, so that when there are nodes joining and exiting operations in the local network, the backup data of the external domain is not required to be adjusted, and the extra-domain backup data needs to be adjusted only when the resource itself is changed. This can effectively reduce the overhead of backup data adjustment and improve system efficiency and performance.
  • the method of the present invention and the peer-to-peer network perform extra-domain backup of resource information to better ensure the reliability and availability of data; and additionally perform extra-domain backup for resource information, so that nodes in the local network join and exit operations.
  • the backup data of the external domain is not required to be adjusted, the extra-domain backup data needs to be adjusted only when the resource itself is changed. This can effectively reduce the overhead caused by the adjustment of the backup data and improve the efficiency and performance of the system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明提供一种基于对等网络的资源信息备份操作方法,该方法包括:发起节点向备份域的连接节点发送域外备份节点确定请求,所述域外备份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息,所述备份域的连接节点与所述主节点具有不同的域标识;所述备份域的连接节点根据所述资源全局标识符信息及域外备份规则确定域外备份节点,并将所述域外备份节点的路由信息发送给所述发起节点;所述发起节点根据所述路由信息向所述域外备份节点发送域外备份操作请求,所述域外备份节点根据所述域外备份操作请求进行相应的处理。本发明方法及对等网络,可以更好的保证数据的可靠性和可用性。

Description

基于对等网络的资源信息备份操作方法及对等网络
技术领域
本发明涉及对等网络, 尤其涉及一种基于对等网络的资源信息备份操作 方法及对等网络。
背景技术
对等 (Peer to Peer, P2P ) 网络作为现有互联网 (Internet ) 的覆盖网络 ( Overlay Network ) , 通过系统之间的对等交换来共享计算机资源和服务, 以实现分布式、 可靠、 可扩展、 健壮的网络应用模式。 它打破了传统的客户 端 /服务器( Client/Server, C/S)模式,在网络中的每个结点的地位都是对等的。 每个结点既充当服务器, 为其他结点提供服务, 同时也享用其他结点提供的 服务。
随着互联网和宽带接入网的迅速发展, P2P技术得到了大量的应用和研 究, 特别是基于分布式散列表(Distributed Hash Table, 简称 DHT )和叠加网 络( Overlay Network )的全分布式结构化 P2P网络(即 DHT网络)得到业界 的广泛重视。
在 DHT网络中, 某些节点可能正常或异常离开该网络, 这样就容易造成 存储资源的丟失, 为了增强存储资源的可用性, 使其不会因为个别节点的失 效而无法获得, 通常要在 DHT 网络中利用别的节点来为这些资源做一些备 份, 用来备份某个资源的节点叫做这个资源的备份节点。 备份资源使得资源 的可用性得到提高。
现有技术,资源的备份主要是在一个 DHT网络内部进行的。例如在 Chord 中, 节点按照其节点 ID从小到大顺时针组成一个 DHT环, 取某个节点在顺 时针方向遇到的第一个节点作为其后向节点, 由于每个节点维护该节点的后 向节点列表, 因此节点所存储的数据直接备份在其后向节点上, 如果需要存 储多份, 则存储在其后向的多个节点上。 这种备份方式维护比较方便, 但在 一些异常情况下, 这种方法并不能保证资源的可用性。 现有技术有很多对于 DHT网络的内部备份方式进行了改进, 主要是针对两个方面: 一是通过把数 据复制多份, 并按一定规则存放在若干个节点上, 恢复时只需取出一个数据 即可; 另一种是将一个数据分为多个冗余片段, 这样只需取出其中几个片段 便可对数据进行恢复, 来实现其数据的备份, 这种备份方式的存储量比较小, 但维护开销比较大。
由于上述的备份方式只是针对 DHT网络内部进行的,当网络中只有个别 少数节点退出网络时, 这种方式对网络的数据恢复可以工作的很好, 但当网 络中突然有大量节点异常退出时, 则很可能造成部分数据不能恢复, 而这对 于可靠性要求很高的网络来说是不允许的。
发明内容
本发明要解决的技术问题是提供一种基于对等网络的资源信息备份操作 方法及对等网络, 以更好的保证数据的可靠性和可用性。
为解决上述技术问题, 本发明提供一种基于对等网络的资源信息备份操 作方法,所述对等网络的节点具有包括域标识及节点标识的节点全局标识符, 所述对等网络的资源信息具有包括域标识及资源标识的资源全局标识符, 所 述资源信息保存在与所述资源信息具有相同域标识的主节点上, 对所述资源 信息进行备份操作的方法包括:
发起节点向备份域的连接节点发送域外备份节点确定请求, 所述域外备 份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息, 所述 备份域的连接节点与所述主节点具有不同的域标识;
所述备份域的连接节点根据所述资源全局标识符信息及域外备份规则确 定域外备份节点, 并将所述域外备份节点的路由信息发送给所述发起节点; 所述发起节点根据所述路由信息向所述域外备份节点发送域外备份操作 请求, 所述域外备份节点根据所述域外备份操作请求进行相应的处理。
上述方法还包括: 若所述域外备份操作请求是指域外备份请求, 则所述 发起节点是所述资源信息所在的主节点, 所述主节点向所述域外备份节点发 送的域外备份请求中还携带所述资源信息的资源信息内容, 所述域外备份节 点根据所述域外备份请求对所述资源信息内容及资源全局标识符信息进行备 份;
若所述域外备份操作请求是指域外备份查询请求, 则所述发起节点是查 询节点, 所述域外备份节点根据域外备份查询请求中的资源全局标识符信息 向所述查询节点返回要查询的资源信息的资源信息内容;
若所述域外备份操作请求是指域外备份恢复请求, 则所述发起节点是检 测到所述资源信息所在主节点故障的节点, 所述资源全局标识符信息指要恢 复的资源信息的资源全局标识符或资源全局标识符范围, 所述资源全局标识 符或资源全局标识符范围是所述发起节点根据域内资源存放规则确定的, 所 述域外备份节点根据所述域外备份恢复请求向所述发起节点返回所述资源全 局标识符对应的资源信息内容或所述资源全局标识符范围内的资源全局标识 符对应的资源信息内容, 并由发起节点进行保存。
上述方法还包括: 所述对等网络的节点保存有资源信息表、 域内路由表 及域外路由表, 其中所述资源信息表包括资源信息的资源全局标识符及资源 信息内容, 所述域内路由表包括与所述节点具有连接关系的域内节点的节点 全局标识符及路由信息, 所述域外路由表包括与所述节点具有连接关系的域 外节点的节点全局标识符及路由信息, 所述域外备份节点是结合域内路由表 中的节点全局标识符确定的。
为解决上述技术问题, 本发明还提供一种对等网络, 所述对等网络的节 点具有包括域标识及节点标识的节点全局标识符, 所述对等网络的资源信息 具有包括域标识及资源标识的资源全局标识符, 所述对等网络的节点包括接 收模块、 与接收模块连接的域外备份节点确定模块及与所述接收模块及所述 域外备份节点确定模块连接的处理模块, 其中:
接收模块, 用于接收发起节点发送的域外备份节点确定请求, 所述域外 备份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息, 所 述备份域的连接节点与所述主节点具有不同的域标识; 还用于接受所述发起 节点发送的域外备份操作请求;
域外备份节点确定模块, 用于根据所述资源全局标识符信息及域外备份 规则确定域外备份节点; 处理模块, 用于根据确定的域外备份节点向所述发起节点发送所述域外 备份节点的路由信息; 还用于根据所述域外备份操作请求进行相应处理。
此外, 若所述域外备份操作请求是指域外备份请求, 则所述发起节点是 所述资源信息所在的主节点, 所述域外备份操作请求中还携带所述资源信息 的资源信息内容, 所述处理模块用于根据所述域外备份操作请求对所述资源 信息内容及资源全局标识符信息进行备份;
若所述域外备份操作请求是指域外备份查询请求, 则所述发起节点是查 询节点, 所述处理模块用于根据所述资源全局标识符信息向所述查询节点返 回要查询的资源信息的资源信息内容;
若所述域外备份操作请求是指域外备份恢复请求, 则所述发起节点是检 测到所述资源信息所在主节点故障的节点, 所述域外备份恢复请求中的资源 全局标识符信息指要恢复的资源信息的资源全局标识符或资源全局标识符范 围 , 所述资源信息的资源全局标识符或资源全局标识符范围是所述发起节点 根据域内资源存放规则确定的, 所述处理模块用于根据所述域外备份恢复请 求向所述发起节点返回所述资源全局标识符对应的资源信息内容或所述资源 全局标识符范围内的资源全局标识符对应的资源信息内容。
此外, 所述节点还包括与所述域外备份节点确定模块及处理模块连接的 存储模块, 用于存储资源信息表、 域内路由表及域外路由表, 其中所述资源 信息表包括资源信息的资源全局标识符及资源信息内容, 所述域内路由表包 括与所述节点具有连接关系的域内节点的节点全局标识符及路由信息, 所述 域外路由表包括与所述节点具有连接关系的域外节点的节点全局标识符及路 由信息, 所述域外备份节点是结合域内路由表中的节点全局标识符确定的。
为解决上述技术问题, 本发明还提供一种基于对等网络的资源信息备份 操作方法, 所述对等网络的节点具有包括域标识及节点标识的节点全局标识 符,所述对等网络的资源信息具有包括域标识及资源标识的资源全局标识符, 所述资源信息保存在与所述资源信息具有相同域标识的主节点上, 对所述资 源信息进行备份操作的方法包括:
发起节点向备份域的连接节点发送域外备份操作请求, 所述域外备份节 点操作请求中包括要备份操作的资源信息的资源全局标识符, 所述备份域的 连接节点与所述主节点具有不同的域标识;
所述备份域的连接节点根据所述资源全局标识符及域外备份规则确定域 外备份节点, 并向所述域外备份节点转发所述域外备份操作请求, 其中携带 发起节点的路由信息;
所述域外备份节点根据所述域外备份操作请求进行相应处理。
上述方法还包括: 所述的域外备份操作请求, 是域外备份请求、 备份查 询请求或域外备份恢复请求; 若域外备份操作请求是域外备份请求, 则请求 中还携带要备份的资源信息内容。
为解决上述技术问题, 本发明还提供一种对等网络, 所述对等网络的节 点具有包括域标识及节点标识的节点全局标识符, 并保存有域内其他节点的 节点全局标识符, 所述对等网络的资源信息具有包括域标识及资源标识的资 源全局标识符, 所述资源信息保存在与其有相同域标识的节点上, 同时备份 在与其有不同域标识的域外备份节点上, 所述域外备份节点是根据所述资源 信息的资源全局标识符、 备份域的节点全局标识符及域外备份规则确定的。
上述方法还包括: 所述资源信息同时备份在与其有相同域标识的其他节 点上。
相较于现有技术, 本发明方法及对等网络, 对资源信息进行域外备份, 更好的保证了数据的可靠性和可用性; 另外针对资源信息进行域外备份, 这 样在本域网络中有节点加入退出等操作时, 不需要对外域的备份数据进行调 整, 只有当该资源本身有更改时才需要对域外备份数据进行调整, 这样可以 有效的减少备份数据调整带来的额外开销, 提高系统的效率和性能。
附图概述
图 1为本发明 P2P网络架构的示意图;
图 2为本发明中节点和资源信息的全局标识符示意图;
图 3为本发明中节点存储的内容示意图;
图 4为本发明资源信息备份操作方法的流程示意图; 图 5为本发明方法应用实例 1资源信息的备份过程示意图; 图 6为本发明方法应用实例 2备份资源信息的查找示意图;
图 7为本发明方法应用实例 3新节点加入过程的示意图;
图 8为本发明方法应用实例 4利用域外备份数据对资源信息进行恢复示 意图;
图 9为本发明对等网络节点的模块结构示意图。
本发明的较佳实施方式
本发明资源备份方法是基于地域优化的 P2P网络架构实现的, 下面对此 网络架构进行简要介绍:
在此网络架构中整个 P2P网络按照地理位置划分为若干区域, 并确定每 个区域的区域标识(即域标识, 简称域 ID ) , 拥有相同域标识的节点构成一 个域,域内节点按照一定的 DHT算法构成一个子 P2P网络, 区域内的每个节 点按照一定规则与其他域外保持联系 (见图 1 ) 。
网络中每个节点(或资源)拥有全网唯一的节点(或资源)全局标识符, 如图 2所示, 此全局标识符由域标识和节点 (或资源)标识共同组成, 节点 和资源根据它所在地域决定它的域标识, 节点 (或资源)标识通过对节点 (或 资源) 的某些信息如节点 IP、 端口号 (或资源名称)进行哈希算法得到。 网 络中节点 (或资源) 的插入、 查找等均才艮据节点 (或资源) 的全局标识符进 行。 在此网络中, 每个资源存放在本域(即域标识与其自身相同的域) 的节 点上, 且此节点的全局标识符应和该资源的资源全局标识符相等或相近(一 般取其后向最近) 。
节点标识是通过对节点的特征信息进行哈希运算得到, 节点的特征信息 指该节点的唯一性信息, 如 IP地址、 网卡号、 或机器序号, 哈希算法的选取 可以是任意的;
资源标识是通过对资源信息的特征信息进行哈希算法得到的, 所说资源 的特征信息是能标识该资源的信息, 如资源的具体内容、 用户名、 用户数据 的名称等, 哈希算法的选取可以是任意的。 每个节点中所需存储的内容如图 3所示, 包括路由信息表、 资源信息表 以及域 ID映射表。其中路由信息表包括节点所在域的本域的域内路由表和与 本节点保持联系的域外节点路由信息的域外路由表; 资源信息表包括该节点 所存放的资源信息以及其存放的备份资源信息。 这里所说的备份资源信息包 括域外资源信息的备份及域内资源信息的备份。
资源信息表中存放的资源信息包括资源信息的资源全局标识符及资源信 息内容。
在实际网络中, 大量网络节点的异常宕机往往发生在某个临近地域上, 为了更好的保证数据的可用性, 本发明在对资源进行备份时, 不仅在本域进 行数据备份, 而且在域外 (域 ID与资源的域 ID不同的域)进行备份, 从而 在保持域内备份优点的同时满足异地容灾的要求, 保证数据的可恢复性。
以下对本发明中的一些定义进行说明: 主节点: 资源信息按照存储规则在本域中被保存的节点; 具体地, 存储 规则为:资源存放在某个节点上,这个节点的节点 ID和此资源信息的资源 ID 最相近且大于等于资源 ID,如果不存在比该资源 ID更大的节点 ID,则存储在 该域内节点 ID最小的节点上。
本域: 主节点所在的域;
本域备份节点: 按照一定的本域备份规则备份主节点资源信息的本域节 点; 具体地, 本域备份规则可以为: 主节点的后向节点作为本域备份节点。
节点根据节点 ID的大小按一定次序(一般为升序)排列成一个大的圓环, 称为 DHT环(如图 5所示), 取某个节点其顺时针方向遇到的第一个节点作 为其后向节点, 其逆时针方向遇到第一个节点为其前向节点, 每个节点均有 一个前向节点和一个后向节点。 备份域: 用于本域数据备份的域, 备份域的选取规则一般为选择规模和 本域相当或负载较轻的域作为备份域; 备份域的选取可以通过人为指定本域 和备份域的映射表, 也可以利用一些算法根据域的规模选择本域的备份域。
域外备份节点: 按照一定的域外备份规则备份资源信息的域外节点。 选 择域外备份节点时,可以根据本域资源信息的资源 ID将所述资源备份到备份 域的相应的备份节点上,例如域外备份节点可以是在备份域中节点 ID与该资 源 ID相等或最接近的节点, 且大于等于该资源 ID的第一个节点; 当有数据 加入、 删除或更新时对备份域的备份数据进行相应更改。
本发明根据资源标识对资源信息进行备份, 当本域节点发生变化时, 外 域节点上存放的相应备份数据不需要调整。 只有资源信息改变 (如资源信息加 入、 删除或更改) 时需要在外域进行相应的改变。
如图 4所示, 本发明基于对等网络的资源信息备份操作的方法包括: 步骤 401 : 发起节点向备份域的连接节点发送域外备份节点确定请求; 所述域外备份节点确定请求中包括要备份操作的资源信息的资源全局标 识符信息, 所述备份域的连接节点与所述主节点具有不同的域标识;
这里所说的资源全局标识符信息可以是确定的资源全局标识符或资源全 局标识符范围。
步骤 402: 所述备份域的连接节点根据所述资源全局标识符信息及域外 备份规则确定域外备份节点, 并将所述域外备份节点的路由信息发送给所述 发起节点;
步骤 403: 所述发起节点根据所述路由信息向所述域外备份节点发送域 外备份操作请求, 所述域外备份节点根据所述域外备份操作请求进行相应的 处理。
若所述域外备份操作请求是指域外备份请求, 则所述发起节点是所述资 源信息所在的主节点, 所述主节点向所述域外备份节点发送的域外备份请求 中还携带所述资源信息的资源信息内容, 所述域外备份节点根据所述域外备 份请求对所述资源信息内容及资源全局标识符信息进行备份;
若所述域外备份操作请求是指域外备份查询请求, 则所述发起节点是查 询节点, 所述域外备份节点根据域外备份查询请求中的资源全局标识符信息 向所述查询节点返回要查询的资源信息的资源信息内容;
若所述域外备份操作请求是指域外备份恢复请求, 则所述发起节点是检 测到所述资源信息所在主节点故障的节点, 所述资源全局标识符信息指要恢 复的资源信息的资源全局标识符或资源全局标识符范围, 所述资源全局标识 符或资源全局标识符范围是所述发起节点根据域内资源存放规则确定的, 所 述域外备份节点根据所述域外备份恢复请求向所述发起节点返回所述资源全 局标识符对应的资源信息内容或所述资源全局标识符范围内的资源全局标识 符对应的资源信息内容, 并由发起节点进行保存。
以上第一方法中, 是由发起节点根据备份域连接节点返回的路由信息向 域外备份节点发送域外备份操作请求的, 当然也可以由备份域连接节点向域 外备份节点转发域外备份操作请求, 其中携带发起节点的路由信息, 以便域 外备份节点处理后向发起节点返回处理结果, 第二方法具体包括:
步骤 A: 发起节点向备份域的连接节点发送域外备份操作请求, 所述域 外备份节点操作请求中包括要备份操作的资源信息的资源全局标识符信息, 所述备份域的连接节点与所述主节点具有不同的域标识;
步骤 B: 所述备份域的连接节点根据所述资源全局标识符信息及域外备 份规则确定域外备份节点, 并向所述域外备份节点转发所述域外备份操作请 求, 其中携带要操作的资源信息的资源全局标识符信息及发起节点的路由信 息;
步骤 C: 所述域外备份节点根据所述域外备份操作请求进行相应处理并 根据所述路由信息向所述发起节点返回处理结果。
以上所述的域外备份操作请求, 是域外备份请求、 备份查询请求或域外 备份恢复请求; 若域外备份操作请求是域外备份请求, 则请求中还需要携带 要备份的资源信息内容。
当本域主节点 (存放该资源原始数据的节点)和本域备份节点 (本域中 用于数据备份的节点) 均由于某种异常而导致数据不可获得时可以通过备份 域来完成数据的查找和恢复。
具体地, 数据查找时根据资源信息的全局标识符进行, 当本域内未发现 所需数据时, 则向备份域查找, 查找请求为查询备份数据的请求, 该请求与 直接数据查询请求的查询方式不同;
当本域的主节点和本域备份节点均异常宕机时, 本域的某个节点 (根据 域内釆用的 DHT算法和备份的方法而定)检测到这种情况发生, 则主动向备 份域发出数据恢复请求, 请求中包括需要恢复的资源信息的全局资源标识符 范围。
在本域内釆用某种域内备份的方法对本域数据进行备份, 域内的备份方 法可以釆用单个 DHT网络的备份方法, 备份方法的选取可以是任意的。
现在结合具体的实施例对本发明作进一步的说明:
图 5、 6、 7、 8中每个节点为一个归属用户服务器(Home Subscriber Server,
HSS ) , HSS为支持用于处理调用 /会话的 IMS网络实体的主要用户数据库。 它包含用户配置文件, 执行用户的身份验证和授权, 并可提供有关用户物理 位置的信息。 在此实施例中, 每个 HSS (以下简称节点) 中保存了部分用户 的用户数据, 称为资源信息。
为简化起见, 图中给出了两个域: 域 A与域 B, 其中节点 N7~N46属于 域 A, 节点 n3〜! 153属于域 B, 节点的编号即为其节点 ID。 其中域 B作为域 A 的备份域。 在实际网络中按照域的规模或域的负载状况选择备份域, 当备份 域选定之后, 当新节点加入时就会指定其备份域的域 ID, A域的节点都会保 存这个信息。 域 A同时也作为其他域的备份域。 为了简化, 其他域在图中并 未表示。
图中每个节点在域内釆用 One-hop DHT算法(即域内每个节点均保存一 份域内的全局路由表, 从而和域内的其他节点保持全连接) , 此外每个节点 还保存一份域外路由表, 此域外路由表包含域外的部分节点的路由信息。
应用实例 1 : 数据备份
在本发明中主要针对域外备份, 对于域内备份在此釆用最简单的形式, 即选取主节点的后向节点作为备份节点, 备份节点用于备份主节点上的全部 资源信息 (当有新数据加入或数据更新时保持对备份数据的及时更新) 。
该实施例中的域外备份规则是: 备份资源信息的域外节点为具有这样节 点标识的节点, 其节点标识等于或大于所述资源信息的资源标识, 且最接近 所述资源信息的资源标识。
如图 5所示, 数据备份的步骤如下:
步骤 501 : 当有新数据加入或有数据更新时, 主节点 (节点 N7 )对新数 据 (其资源 ID=5 )进行存储;
步骤 502: 节点 N7给其后向节点 (节点 N12 )发送备份请求对新数据进 行备份;
步骤 503: 节点 N12对新数据进行备份并返回应答;
步骤 504: 节点 N7对新加入的资源信息进行域外备份, 它向对应备份域 的连接节点 (例如节点 n53)发送域外备份请求,请求中包括要备份的资源信息 的资源全局标识符;
其中备份域的连接节点是随机选择的。
步骤 505:节点 n53从请求中包含的资源全局标识符中提取出资源 ID(资 源 ID=5 ) , 通过查找自己的域内路由表得到此资源信息应备份到的节点(即 等于或大于此资源 ID且最接近的节点, 称为域外备份节点) , 在此为节点 n9, 节点 n53把节点 n9的路由信息发送给发起备份请求的节点 N7;
步骤 506: 节点 N7向节点 n9发出域外备份请求, 请求中包括资源信息 的资源全局标识符和资源信息内容;
步骤 507: 节点 n9根据收到的请求对资源信息进行备份, 并返回备份成 功应答。
应用实例 2: 域外备份数据的查找
数据查找分为域内数据查找和域外备份数据的查找, 在此只给出域外备 份数据的查找过程, 对于域内数据的查找可以直接通过查询资源信息的全局 标识符按照数据存储规则在本域内查询得到 (对于域外备份数据的查找只有 当本域数据和本域的备份数据均不可用时才会进行) , 如图 6所示, 查询过 程包括:
步骤 601 : 查询节点 N23才艮据需查询数据的资源全局标识符查找自己的 域内路由表, 利用资源全局标识符中的资源 ID (例如资源 ID=6 )得到数据的 主节点 (在此为节点 N7 ) 的路由信息, 并向节点 N7发出查询请求, 请求中 包括需要查询的数据的资源全局标识符;
步骤 602: 由于节点 N7异常宕机或其他情况造成请求失败, 节点 N23 向域内的备份节点 (节点 N12 )发出查询备份数据请求;
步骤 603: 节点 N12也发生某种异常, 查询请求失败, 节点 N23向备份 域的其已知的某个节点 (例如节点 n32)发出域外备份查询请求,请求中包括需 查找的数据的资源全局标识符; (这里的查询备份数据的请求和直接进行数 据查询的请求是不同的, 这里请求中包括该资源的全局标识符是因为该备份 域可能还存有其他域的备份数据) ;
步骤 604: 节点 n32按照资源全局标识符和域外备份规则来确定该资源 的域外备份节点,并查找自己的域内路由表得到域外备份节点 (在此为节点 n9) 的路由信息, 并向节点 n9转发域外备份查询请求, 转发的查询请求中应该携 带 N23的路由信息;
步骤 605: 节点 n9向节点 N23返回所需查询的资源信息
域外备份数据查找只发生在本域的数据主节点和备份节点均查询失败的 情况, 这种情况通常是由于域内大量节点异常宕机所造成的, 这样通过域外 备份数据的查找仍然能实现数据的正常获得。
应用实例 3: 节点加入 /退出时的数据迁移
当有新节点 (例如节点 N20 )加入的时候, 需要从后向节点复制的数据 包括三部分, 如图 7所示, 具体复制步骤如下:
步骤 701 : 新节点作为主节点负责保存的资源信息 (资源 ID值处于前 向节点 ID与新节点 ID之间的那部分资源信息),在此为资源 ID为( 16, 20 ] (在此表示该节点作为主节点应负责保存的数据范围,这些数据的节点 ID应 大于 16而小于等于 20; )之间的数据;
步骤 702: 新节点作为域外备份节点所应该负责保存的备份数据 (后向 节点中保存的资源 ID小于或等于本域新节点 ID的备份数据 ) 。 在此假设域 A为域 C的备份域(图未示) , 则此备份资源信息为域 C中资源 ID在 (16, 20 ] 的数据; 步骤 703: 新节点作为域内备份节点所应该负责保存的前向节点的备份 数据(后向节点中保存的本域备份数据), 在此为资源 ID在(12, 16 ]的数 据。
以上几步只需从新加入节点的后向节点 (即 N23 )进行复制即可, 复制 完数据后, 新节点需要进行域内数据备份, 而不需要进行域外数据的备份, 然后该节点按照本域的路由更新方法及时通知其他节点更新路由表。 同理, 域内节点的正常离开是加入的一个逆过程, 离开时, 此节点把负责保存的全 部资源信息移交至其后向节点, 后向节点仅需要做域内备份, 不需要再重新 进行域外备份。 应用实例 4: 数据恢复过程
在此仅考虑域内主节点和备份节点同时出故障的情况, 对于只有其中之 一出故障的情况, 可以通过域内数据恢复机制进行恢复, 而且不需要更改域 外的备份数据。 域内备份节点的后向节点 N16检测出两个以上连续节点发生 故障后 (节点 N16通过与其前向节点之间的保活机制来检测其前向连续多个 节点的故障) , 会根据域外备份的规则从所丟失数据域外备份节点处将资源 信息复制过来, 并重新进行备份。
如图 8所示, 具体步骤如下:
步骤 801 : 域内备份节点的后向节点 N16检测出故障 (假设只有 N7和 N12发生故障)后, 只能够得知域内丟失资源信息的 ID范围为 [失效主节点 的前向节点 ID, 失效主节点 ID] , 失效主节点在此为节点 N7, 所丟失资源信 息为资源 ID位于(46, 7]的资源信息, 即资源 ID大于 46或不大于 7的资源 信息;
该实施例中的域外备份规则是: 备份资源信息的域外节点为具有这样节 点标识的节点, 其节点标识等于或大于所述资源信息的资源标识, 且最接近 所述资源信息的资源标识。
节点 N16需要进行的数据恢复和备份包括:
i、 对域内丟失的数据通过域外备份进行恢复; ii、 把恢复的数据在节点 N16的后向节点进行域内备份;
iii、对节点 N16的前向节点 N12之前备份在节点 N16上的数据进行域内 备份;
iv、 对失效主节点前向节点所存放的数据进行备份;
v、 对失效节点所负责的域外的备份数据进行重新备份;
步骤 802: 节点 N16用失效主节点前向节点 N46的节点 ID=46在备份域 中查询 (可以通过其域外路由表中的所保存的域 B中的节点的路由信息得到 节点 n43的路由信息) , 找到此节点 ID对应的后向节点为节点 n53; 再用节 点 N7的节点 ID=7在备份域中查询, 找到其后向节点 n9; 节点 N16向节点 n53请求查询节点 ID位于 (53 , 7 )之间的节点的路由信息。 则原域内丟失 的资源信息为:
i、 节点 N53中备份的资源 ID值在 (46, 53] 之间的数据;
ii、 节点 n9中备份资源 ID值在(3 , 7] 之间的数据;
iii、节点 n53和节点 n9之间的节点所备份的所有数据, 即节点 n3上备份 的所有数据,大于 53及小于 3的数据;
步骤 803: 节点 N16将所需数据从域外中拷贝过来, 此时节点 N16成为 这些资源信息的主节点;
步骤 804: 节点 N16重新对刚恢复的数据以及节点 N16的失效前向节点 存放在 N16上的域内备份数据进行域内备份, 即把恢复的数据以及域内备份 的数据备份到其后向节点 N23上;
步骤 805: 节点 N16将节点 N46所存放的数据进行备份;
步骤 806: 所丟失的域外备份数据的范围为: (失效主节点的前向节点 ID, 节点 N16前向节点 ID] , 即资源 ID位于(46, 12]范围的域外备份数据。 如果这些备份数据涉及到多个域, 节点 N16需要到多个域分别对所需数据进 行重新备份。 具体过程和对失效节点的数据恢复相同, 不赘述。
如果需要进一步保证数据的可靠性, 可以考虑在域外进行多个备份, 例 如除了在域外的备份节点进行备份一次外, 再在该域外备份节点的后向节点 备份一份该资源。 以上过程均不必重新进行域外备份,域外中的备份数据不必做任何改动。 为实现以上第一方法, 本发明还提供了一种网络架构, 并对节点进行了 功能模块划分, 如图 9所示, 所述对等网络的节点包括接收模块、 与接收模 块连接的域外备份节点确定模块、 与所述接收模块及所述域外备份节点确定 模块连接的处理模块, 及所述节点还包括与所述域外备份节点确定模块及处 理模块连接的存储模块, 其中:
接收模块, 用于接收发起节点发送的域外备份节点确定请求, 所述域外 备份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息, 所 述备份域的连接节点与所述主节点具有不同的域标识; 还用于接受所述发起 节点发送的域外备份操作请求;
域外备份节点确定模块, 用于根据所述资源全局标识符信息及域外备份 规则确定域外备份节点;
处理模块, 用于根据确定的域外备份节点向所述发起节点发送所述域外 备份节点的路由信息; 还用于根据所述域外备份操作请求进行相应处理。
若所述域外备份操作请求是指域外备份请求, 则所述发起节点是所述资 源信息所在的主节点, 所述域外备份操作请求中还携带所述资源信息的资源 信息内容, 所述处理模块用于根据所述域外备份操作请求对所述资源信息内 容及资源全局标识符信息进行备份;
若所述域外备份操作请求是指域外备份查询请求, 则所述发起节点是查 询节点, 所述处理模块用于根据所述资源全局标识符信息向所述查询节点返 回要查询的资源信息的资源信息内容;
若所述域外备份操作请求是指域外备份恢复请求, 则所述发起节点是检 测到所述资源信息所在主节点故障的节点, 所述域外备份恢复请求中的资源 全局标识符信息指要恢复的资源信息的资源全局标识符或资源全局标识符范 围 , 所述资源信息的资源全局标识符或资源全局标识符范围是所述发起节点 根据域内资源存放规则确定的, 所述处理模块用于根据所述域外备份恢复请 求向所述发起节点返回所述资源全局标识符对应的资源信息内容或所述资源 全局标识符范围内的资源全局标识符对应的资源信息内容。 存储模块, 用于存储资源信息表、 域内路由表及域外路由表, 其中所述 资源信息表包括资源信息的资源全局标识符及资源信息内容, 所述域内路由 表包括与所述节点具有连接关系的域内节点的节点全局标识符及路由信息, 所述域外路由表包括与所述节点具有连接关系的域外节点的节点全局标识符 及路由信息, 所述域外备份节点是结合域内路由表中的节点全局标识符确定 的。
本发明方法及对等网络, 域内的数据备份可以保证在网络中少量节点正 常或异常离开网络时的数据的可恢复性, 本发明在保持域内备份的优点的同 时, 也实现了异地容灾, 更好的保证了数据的可靠性和可用性。 另外本发明 针对资源信息进行域外备份, 这样在本域网络中有节点加入退出等操作时, 不需要对外域的备份数据进行调整, 只有当该资源本身有更改时才需要对域 外备份数据进行调整, 这样可以有效的减少备份数据调整带来的额外开销, 提高系统的效率和性能。
工业实用性 本发明方法及对等网络, 对资源信息进行域外备份, 更好的保证了数据 的可靠性和可用性; 另外针对资源信息进行域外备份, 这样在本域网络中有 节点加入退出等操作时, 不需要对外域的备份数据进行调整, 只有当该资源 本身有更改时才需要对域外备份数据进行调整, 这样可以有效的减少备份数 据调整带来的额外开销, 提高系统的效率和性能。

Claims

权 利 要 求 书
1、 一种基于对等网络的资源信息备份操作方法, 其特征在于: 所述对 等网络的节点具有包括域标识及节点标识的节点全局标识符, 所述对等网络 的资源信息具有包括域标识及资源标识的资源全局标识符, 所述资源信息保 存在与所述资源信息具有相同域标识的主节点上, 所述方法包括:
发起节点向备份域的连接节点发送域外备份节点确定请求, 所述域外备 份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息, 所述 备份域的连接节点与所述主节点具有不同的域标识;
所述备份域的连接节点根据所述资源全局标识符信息及域外备份规则确 定域外备份节点, 并将所述域外备份节点的路由信息发送给所述发起节点; 所述发起节点根据所述路由信息向所述域外备份节点发送域外备份操作 请求, 所述域外备份节点根据所述域外备份操作请求进行相应的处理。
2、 如权利要求 1所述的方法, 其中:
若所述域外备份操作请求是指域外备份请求, 则所述发起节点是所述资 源信息所在的主节点,
所述向所述域外备份节点发送域外备份操作请求的步骤还包括: 所述主 节点向所述域外备份节点发送的所述域外备份请求中还携带所述资源信息的 资源信息内容,
所述域外备份节点进行相应的处理的步骤为: 所述域外备份节点根据所 述域外备份请求对所述资源信息内容及资源全局标识符信息进行备份;
若所述域外备份操作请求是指域外备份查询请求, 则所述发起节点是查 询节点, 所述域外备份节点进行相应的处理的步骤为: 所述域外备份节点根 据所述域外备份查询请求中的资源全局标识符信息向所述查询节点返回要查 询的资源信息的资源信息内容;
若所述域外备份操作请求是指域外备份恢复请求, 则所述发起节点是检 测到所述资源信息所在主节点故障的节点, 所述资源全局标识符信息指要恢 复的资源信息的资源全局标识符或资源全局标识符范围, 所述资源全局标识 符或资源全局标识符范围是所述发起节点根据域内资源存放规则确定的, 所述域外备份节点进行相应的处理的步骤为: 所述域外备份节点根据所 述域外备份恢复请求向所述发起节点返回所述资源全局标识符对应的资源信 息内容或所述资源全局标识符范围内的资源全局标识符对应的资源信息内 容, 并由发起节点进行保存。
3、 如权利要求 1 所述的方法, 其中, 所述对等网络的节点保存有资源 信息表、 域内路由表及域外路由表, 所述资源信息表包括资源信息的资源全 局标识符及资源信息内容, 所述域内路由表包括与所述节点具有连接关系的 域内节点的节点全局标识符及路由信息, 所述域外路由表包括与所述节点具 有连接关系的域外节点的节点全局标识符及路由信息,
所述确定域外备份节点的步骤中, 所述域外备份节点是结合域内路由表 中的节点全局标识符确定的。
4、 一种用于对等网络的的节点, 其特征在于: 所述节点具有包括域标 识及节点标识的节点全局标识符, 所述对等网络的资源信息具有包括域标识 及资源标识的资源全局标识符, 所述节点在对等网络中用作发起节点、 主节 点、 连接节点、 域外备份节点;
所述节点包括接收模块、 与接收模块连接的域外备份节点确定模块及与 所述接收模块及所述域外备份节点确定模块连接的处理模块,
接收模块设置为: 接收发起节点发送的域外备份节点确定请求, 所述域 外备份节点确定请求中包括要备份操作的资源信息的资源全局标识符信息 , 备份域的连接节点与主节点具有不同的域标识; 还设置为: 接受所述发起节 点发送的域外备份操作请求;
域外备份节点确定模块设置为: 根据所述资源全局标识符信息及域外备 份规则确定域外备份节点;
处理模块设置为: 根据确定的域外备份节点向所述发起节点发送所述域 外备份节点的路由信息; 还设置为: 根据所述域外备份操作请求进行相应处 理。
5、 如权利要求 4所述的节点, 其中:
所述处理模块设置为: 若所述域外备份操作请求是指域外备份请求, 所 述发起节点是所述资源信息所在的主节点, 所述域外备份操作请求中还携带 所述资源信息的资源信息内容时, 根据所述域外备份操作请求对所述资源信 息内容及资源全局标识符信息进行备份;
或者, 所述处理模块设置为: 若所述域外备份操作请求是指域外备份查 询请求, 所述发起节点是查询节点时, 根据所述资源全局标识符信息向所述 查询节点返回要查询的资源信息的资源信息内容;
或者, 所述处理模块设置为: 若所述域外备份操作请求是指域外备份恢 复请求, 所述发起节点是检测到所述资源信息所在主节点故障的节点, 所述 域外备份恢复请求中的资源全局标识符信息指要恢复的资源信息的资源全局 标识符或资源全局标识符范围, 所述资源信息的资源全局标识符或资源全局 标识符范围是所述发起节点根据域内资源存放规则确定的时, 根据所述域外 备份恢复请求向所述发起节点返回所述资源全局标识符对应的资源信息内容 或所述资源全局标识符范围内的资源全局标识符对应的资源信息内容。
6、 如权利要求 4所述的节点, 其中: 所述节点还包括与所述域外备份 节点确定模块及处理模块连接的存储模块, 所述存储模块设置为: 存储资源 信息表、 域内路由表及域外路由表, 所述资源信息表包括资源信息的资源全 局标识符及资源信息内容, 所述域内路由表包括与所述节点具有连接关系的 域内节点的节点全局标识符及路由信息, 所述域外路由表包括与所述节点具 有连接关系的域外节点的节点全局标识符及路由信息, 所述域外备份节点是 结合域内路由表中的节点全局标识符确定的。
7、 一种基于对等网络的资源信息备份操作方法, 其特征在于: 所述对 等网络的节点具有包括域标识及节点标识的节点全局标识符, 所述对等网络 的资源信息具有包括域标识及资源标识的资源全局标识符, 所述资源信息保 存在与所述资源信息具有相同域标识的主节点上, 所述方法包括:
发起节点向备份域的连接节点发送域外备份操作请求, 所述域外备份节 点操作请求中包括要备份操作的资源信息的资源全局标识符, 所述备份域的 连接节点与所述主节点具有不同的域标识;
所述备份域的连接节点根据所述资源全局标识符及域外备份规则确定域 外备份节点, 并向所述域外备份节点转发所述域外备份操作请求, 所述域外 备份操作请求携带发起节点的路由信息;
所述域外备份节点根据所述域外备份操作请求进行相应处理。
8、 如权利要求 7 所述的方法, 其中: 所述的域外备份操作请求, 是域 外备份请求、 备份查询请求或域外备份恢复请求; 若域外备份操作请求是域 外备份请求, 则所述域外备份请求中还携带要备份的资源信息内容。
9、 一种用于对等网络的节点, 其特征在于: 所述节点在对等网络中用 作包括主节点和域外备份节点的节点; 所述节点具有包括域标识及节点标识 的节点全局标识符, 并保存有域内主节点外的其他节点的节点全局标识符, 所述对等网络的资源信息具有包括域标识及资源标识的资源全局标识符, 所 述资源信息保存在与主节点有相同域标识的节点上, 同时备份在与主节点有 不同域标识的域外备份节点上, 所述域外备份节点是根据所述资源信息的资 源全局标识符、 备份域的节点全局标识符及域外备份规则确定的。
10、 如权利要求 9所述的节点, 其中: 所述资源信息同时备份在与主节 点有相同域标识的其他节点上。
11、 一种对等网络, 其包括如权利要求 4-6中任一所述的节点。
12、 一种对等网络, 其包括如权利要求 9或 10所述的节点。
PCT/CN2010/071617 2009-04-08 2010-04-07 基于对等网络的资源信息备份操作方法及对等网络 WO2010115373A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200910133724.2 2009-04-08
CN200910133724.2A CN101860474B (zh) 2009-04-08 2009-04-08 基于对等网络的资源信息处理方法及对等网络

Publications (1)

Publication Number Publication Date
WO2010115373A1 true WO2010115373A1 (zh) 2010-10-14

Family

ID=42935648

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/071617 WO2010115373A1 (zh) 2009-04-08 2010-04-07 基于对等网络的资源信息备份操作方法及对等网络

Country Status (2)

Country Link
CN (1) CN101860474B (zh)
WO (1) WO2010115373A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017220148A (ja) * 2016-06-10 2017-12-14 富士通株式会社 情報管理プログラム、情報管理方法、及び情報管理装置
CN105472060B (zh) * 2016-01-25 2018-12-11 互联天下科技发展(深圳)有限公司 一种Kademlia网络的节点标识符生成方法
CN114968667A (zh) * 2022-05-30 2022-08-30 江苏安超云软件有限公司 一种备份管理方法及系统

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012003623A1 (zh) * 2010-07-05 2012-01-12 中兴通讯股份有限公司 基于对等网络的资源信息处理方法、对等网络及节点
CN102739707B (zh) * 2011-04-07 2017-08-25 中兴通讯股份有限公司 一种分布式关键字表网络及对其进行操作的方法
CN102868617A (zh) * 2011-07-05 2013-01-09 中兴通讯股份有限公司 一种p2p网络路由方法及系统
CN102891872B (zh) * 2011-07-20 2018-02-09 中兴通讯股份有限公司 一种对等网络中数据存储和查询的方法及系统
CN102572602B (zh) * 2012-01-16 2016-06-29 电子科技大学 在p2p直播流系统中基于dht的分布式索引实现方法
CN103023805A (zh) * 2012-11-22 2013-04-03 北京航空航天大学 一种MapReduce系统
CN103049572B (zh) * 2013-01-04 2015-08-05 福建星网视易信息系统有限公司 适用于有时效保证通讯系统的点对点通讯的文件检索方法
CN103841205A (zh) * 2014-03-14 2014-06-04 北京荣之联科技股份有限公司 一种分布式集群互联方法和装置
CN104113606B (zh) * 2014-08-02 2018-04-10 成都极驰科技有限公司 一致性动态均衡的分布式元数据节点架构的工作方法
CN105550347B (zh) * 2015-12-25 2020-07-14 网易(杭州)网络有限公司 数据处理方法及装置
CN106953804B (zh) * 2016-01-07 2020-08-18 中国移动通信集团海南有限公司 被叫用户的路由信息查询方法及装置
CN108337170B (zh) * 2018-01-30 2021-08-17 浙江省公众信息产业有限公司 分布式资源查找方法和系统
CN108494853B (zh) * 2018-03-22 2021-01-05 广州视源电子科技股份有限公司 一种海量设备状态自维护方法及其装置和系统
CN108712470B (zh) * 2018-04-20 2021-06-29 南京天溯自动化控制系统有限公司 一种路由数据同步方法及装置、计算机装置和存储介质
CN108848191A (zh) * 2018-07-26 2018-11-20 深圳市元征科技股份有限公司 一种节点连接方法及相关设备
CN109639582A (zh) * 2018-12-07 2019-04-16 大科数据(深圳)有限公司 一种基于分布式哈希表的网络节点处理方法及装置
CN114826891A (zh) 2019-05-20 2022-07-29 华为技术有限公司 Tsn中控制器间通信的方法、装置及系统
CN110224931B (zh) * 2019-06-20 2021-09-24 深圳市美兆环境股份有限公司 信息传输方法、装置和计算机设备
CN110602244B (zh) * 2019-09-26 2020-11-03 重庆紫光华山智安科技有限公司 分布式存储系统消息交互方法、节点及分布式存储系统
CN111711654B (zh) * 2020-05-14 2023-03-31 深圳威尔视觉传媒有限公司 P2p通讯连接方法、电子设备及计算机可读存储介质
JP2023533134A (ja) * 2020-05-27 2023-08-02 オッポ広東移動通信有限公司 情報処理方法及び装置、機器、コンピュータ記憶媒体
CN114554567A (zh) * 2020-11-24 2022-05-27 华为技术有限公司 通信的方法及通信装置
CN113746955A (zh) * 2021-11-04 2021-12-03 正链科技(深圳)有限公司 一种分布式网络资源匹配和访问方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1741489A (zh) * 2005-09-01 2006-03-01 西安交通大学 构建多机系统高可用的自愈合逻辑环故障检测与容忍方法
US20060271601A1 (en) * 2005-05-24 2006-11-30 International Business Machines Corporation System and method for peer-to-peer grid based autonomic and probabilistic on-demand backup and restore
CN101079800A (zh) * 2006-08-07 2007-11-28 腾讯科技(深圳)有限公司 一种提高p2p网络稳定性的方法及系统
CN101127653A (zh) * 2006-08-14 2008-02-20 中兴通讯股份有限公司 设置有备份的以太网环网设备及其主设备备份的实现方法
CN101321379A (zh) * 2007-06-08 2008-12-10 华为技术有限公司 CDMA A-Flex网络中确定路由的方法、系统及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101374087A (zh) * 2007-08-20 2009-02-25 华为技术有限公司 一种移动自组网络、节点及其区域划分方法
CN101159966A (zh) * 2007-11-14 2008-04-09 北京邮电大学 在移动环境下的对等网络拓扑匹配方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060271601A1 (en) * 2005-05-24 2006-11-30 International Business Machines Corporation System and method for peer-to-peer grid based autonomic and probabilistic on-demand backup and restore
CN1741489A (zh) * 2005-09-01 2006-03-01 西安交通大学 构建多机系统高可用的自愈合逻辑环故障检测与容忍方法
CN101079800A (zh) * 2006-08-07 2007-11-28 腾讯科技(深圳)有限公司 一种提高p2p网络稳定性的方法及系统
CN101127653A (zh) * 2006-08-14 2008-02-20 中兴通讯股份有限公司 设置有备份的以太网环网设备及其主设备备份的实现方法
CN101321379A (zh) * 2007-06-08 2008-12-10 华为技术有限公司 CDMA A-Flex网络中确定路由的方法、系统及装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105472060B (zh) * 2016-01-25 2018-12-11 互联天下科技发展(深圳)有限公司 一种Kademlia网络的节点标识符生成方法
JP2017220148A (ja) * 2016-06-10 2017-12-14 富士通株式会社 情報管理プログラム、情報管理方法、及び情報管理装置
CN114968667A (zh) * 2022-05-30 2022-08-30 江苏安超云软件有限公司 一种备份管理方法及系统

Also Published As

Publication number Publication date
CN101860474A (zh) 2010-10-13
CN101860474B (zh) 2015-07-22

Similar Documents

Publication Publication Date Title
WO2010115373A1 (zh) 基于对等网络的资源信息备份操作方法及对等网络
WO2010115374A1 (zh) 基于对等网络的资源信息备份操作方法及对等网络
US7859992B2 (en) Router redundancy in data communication networks
US7286468B2 (en) Routing system and method for synchronizing a routing system with peers after failover
JP5227955B2 (ja) 自己統制ネットワークのための分散ハッシュメカニズム
JP6382454B2 (ja) 分散ストレージ及びレプリケーションシステム、並びに方法
CN100452797C (zh) 基于集群路由器结构的高可用分布式边界网关协议系统
US7885180B2 (en) Address resolution request mirroring
JP5016063B2 (ja) 整合性のとれた耐障害性分散ハッシュテーブル(dht)オーバレイネットワーク
CN107465721B (zh) 基于双活架构的全局负载均衡方法和系统及调度服务器
CN111585800B (zh) 基于虚拟私有云的网关节点的配置方法、装置和介质
CN102148850B (zh) 一种集群系统的业务处理方法及集群系统
CN112929285B (zh) 区块链网络的通信优化系统
WO2017193966A1 (zh) 一种存储系统中业务的链路切换方法和存储设备
JP2008533564A (ja) データ管理のための方法および装置
CN110730125B (zh) 一种报文转发方法、装置、双活系统及通信设备
WO2013078852A1 (zh) 分布式哈希表路由表更新方法及节点
WO2012171345A1 (zh) 临时故障时的数据修复方法及分布式缓存系统
WO2009100636A1 (zh) 电信网络用户数据存储管理的方法及装置
JP4335907B2 (ja) ピアツーピアルックアップシステムに対するモビリティチャーン処理のための方法および装置
CN106789634B (zh) 基于链路负载双主环境的静态路由管理方法及系统
WO2008131675A1 (fr) Procédé, nœud de réseau et système pour sauvegarder une ressource dans un réseau p2p structuré
WO2024051672A1 (zh) 一种用于设备单元化调度接入的方法与系统
CN117201575A (zh) 数据发送方法、装置、设备及介质
Xu et al. Building a Robust and Efficient Middleware for Geo-replicated Storage Systems

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: 10761202

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: 10761202

Country of ref document: EP

Kind code of ref document: A1