WO2011113289A1 - 一种业务路由方法和业务网络 - Google Patents

一种业务路由方法和业务网络 Download PDF

Info

Publication number
WO2011113289A1
WO2011113289A1 PCT/CN2010/080478 CN2010080478W WO2011113289A1 WO 2011113289 A1 WO2011113289 A1 WO 2011113289A1 CN 2010080478 W CN2010080478 W CN 2010080478W WO 2011113289 A1 WO2011113289 A1 WO 2011113289A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
cluster
router
global
subnet
Prior art date
Application number
PCT/CN2010/080478
Other languages
English (en)
French (fr)
Inventor
王环
李彦
常恒
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP20100847758 priority Critical patent/EP2541848B1/en
Publication of WO2011113289A1 publication Critical patent/WO2011113289A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63—Routing a service request depending on the request content or context
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/64—Routing or path finding of packets in data switching networks using an overlay routing layer

Definitions

  • the present invention relates to a service routing method and a service network, and specifically relates to a networking method of a service network, a message routing method, a router, and a service network.
  • the embodiment of the present invention provides a networking method for a service network, where the service network includes multiple subnets, and the subnet includes multiple service routers, and the method includes: Determining at least one global service router among the plurality of service routers, the global service routers of the plurality of subnets forming a global peer to peer network layer for routing at least one type of service message between different subnets; The network layer and the plurality of subnets create a service cluster, where the service cluster is used to route messages of the same type of service within the service cluster.
  • an embodiment of the present invention provides a message routing method for a service network, where the method includes: receiving a service interaction message, where the service interaction message includes a destination service address and a destination service type; The destination service type is used to route the service interaction message to the access service router of the destination service by using at least one of a subnet, a global peer network layer, and a service cluster.
  • an embodiment of the present invention provides a router, where the router includes: a subnet routing table, configured to store routing information of a service router in a subnet; and a subnet routing internal routing unit, configured to A routing table within a subnet that routes traffic transmitted within a subnet.
  • the embodiment of the present invention further provides a service network, where the service network includes: multiple subnets, the subnet includes multiple service routers and at least one global service path; a peer-to-peer network layer, formed by a global service router in each of the subnets, for routing messages of at least one type of service between different subnets; at least one service cluster, which is responsible for the same type of service access by multiple subnets
  • the service router is composed of messages for routing the same type of service within the service cluster.
  • the technical solution provided by the embodiment of the present invention is to divide the service network into a local subnet, a service cluster, and a global peer-to-peer network layer, so as to effectively enable the service to access the service network according to the service, which facilitates the management and control of the service.
  • the cross-subnet business interaction uses global peer-to-peer network layer routing to ensure the self-organization capability of the entire service network.
  • Service clusters can set up shortcuts for the same type of service interaction between multiple subnets, reduce the number of route hops for the same type of service interaction, speed up service response, and reduce the burden on the global peer network layer.
  • FIG. 1 is a schematic diagram of a network architecture of a service network according to an embodiment of the present invention
  • FIG. 2 is a functional block diagram of a service router according to an embodiment of the present invention.
  • FIG. 3 is an overall flowchart of a networking method according to an embodiment of the present invention.
  • FIG. 5 is a flowchart 1 of signaling interaction of a route within a subnet according to an embodiment of the present invention
  • FIG. 6 is a second flowchart of signaling interaction of a sub-network route according to an embodiment of the present invention.
  • FIG. 7 is a flowchart 1 of signaling interaction of message routing in a service cluster according to an embodiment of the present invention
  • FIG. 8 is a flowchart 2 of signaling interaction of message routing in a service cluster according to an embodiment of the present invention
  • Signaling interaction flow chart of message routing between subnets
  • FIG. 10 is a flowchart of creating a global peer to peer network layer according to an embodiment of the present invention.
  • FIG. 11 is a flowchart of creating a service cluster according to an embodiment of the present invention.
  • the technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without departing from the inventive scope are the scope of the present invention.
  • the service network (Serv ce Ne twork ) provided by the embodiment of the present invention is a virtual overlay service network established on the physical network layer to provide a unified, secure, and efficient coordinated operation and support environment for all services.
  • the networking method of the service network in the embodiment of the present invention divides the service network into multiple subnets, and each subnet can use different routing protocols according to the requirements and characteristics of the subnet, so that the local service is provided by the local service router.
  • service. Elect or manually designate at least one service router from each subnet to join the global peer-to-peer (P2P, Peer to Peer) network layer (global P2P layer) to route messages between subnets.
  • P2P peer-to-peer
  • Peer to Peer global P2P layer
  • a service type-related service cluster is dynamically created between multiple subnets according to the service type to accelerate the routing of the service and improve the user experience.
  • the service network of the embodiment of the present invention includes a plurality of subnets, a global peer to peer network layer, and at least one service cluster.
  • Subnets can be divided into physical networks or administrative domains.
  • Each subnet includes multiple service routers, and one service router can serve as one node.
  • Each subnet can select a routing protocol according to its own size, business requirements, management requirements and other characteristics.
  • Each subnet can elect or designate at least one service router to join the global P2P layer, that is, each subnet also includes at least one global service router, and the global P2P layer is responsible for routing messages between subnets, that is, can be used in different sub-networks.
  • the network routes at least one type of service.
  • a service router that is responsible for the access of the same type of service between different subnets can form a service cluster.
  • the service cluster is formed by the global P2P layer and the multiple subnets.
  • the interaction message of the same type of service can be directly forwarded by the intra-cluster node. .
  • the foregoing service cluster may include: a cluster head node and a plurality of cluster member nodes, where the cluster head node is configured to save routing information of multiple cluster member nodes, and forward messages between the plurality of cluster member nodes;
  • the cluster member nodes communicate through the cluster head node, for example, the cluster head node is used to forward the message of interaction between the cluster member nodes.
  • the cluster head node provides a query service to each cluster member node, so that each cluster member node directly forwards the message according to the query result, and does not need to be forwarded through the cluster head node.
  • the service network 10 as an example of the embodiment of the present invention includes:
  • the three subnets 12, 14, and 16 are used to be responsible for routing messages in their respective subnets.
  • Each subnet can select its own routing protocol according to its own size, service requirements, management requirements, etc. In actual operation, the subnet
  • the number of nets can be more than three;
  • a global P2P layer 18, from each subnet, can elect or designate at least one service router to be created as a global P2P layer, which is responsible for routing messages between subnets;
  • the service cluster 20 is composed of service routers in the subnet that are responsible for the access of the class A service, and is used to forward the interaction message of the class A service between the nodes in the cluster of the service cluster 20;
  • the service cluster 22 is composed of service routers in the subnet that are responsible for the access of the B-type service, and is used to forward the B-type service interaction messages between the nodes in the cluster of the service cluster 22.
  • the embodiment of the present invention further provides a service router. As shown in FIG. 2, the service router 30 of the embodiment of the present invention includes:
  • the intra-subnet routing table 302 is configured to store routing information of other service routers in the subnet, for example, may include a service network address, an IP address, a port number, and the like of the service router, and the information of the routing table in the subnet may be used along with the information. Routing protocols vary;
  • the intra-subnet routing unit 303 is configured to route the traffic transmitted in the subnet according to the routing table 302 in the subnet, for example, forwarding the destination address to a message in the subnet. Further, the routing unit in the subnet is also used to maintain the topology within the subnet.
  • the service router at the global P2P layer may further include:
  • the global routing table 304 is configured to store routing information of the service router that joins the global P2P layer, where The information including the P2P layer identification ID, IP address, port number, and the like of the service router;
  • the global routing unit 305 is configured to: according to the global routing table 304, route at least one type of service transmitted between the subnets; for example, forward a message between the subnets, and participate in maintaining a topology of the global P2P layer.
  • the routing decision unit 301 is configured to determine a type of the message and a destination address, and determine, according to the type of the message and the destination address, which routing protocol is used to forward the message;
  • the service or the resource has a globally unique addressing in the service network, and the addressing scheme of the service may be designed by the operator.
  • the following addressing format may be used: service identifier, service type, sub- Network identification. Business network identification.
  • the service router also has a globally unique addressing in the service network.
  • the addressing scheme is similar to the addressing scheme of the service, for example: service router identifier. [service type]. subnet identifier. service network identifier, where the service type is optional. .
  • the service and the service router accessing the service may have the same service network address suffix: service type. subnet identifier. service network identifier.
  • the P2P layer ID of the service router is a string formed by a combination of characters and numbers, which can be calculated, such as the algorithm: MD5 (service network address suffix of the service router), where MD5 is a specific hash function.
  • the service router that joins a certain type of service cluster may further include:
  • the service cluster routing table 306 is configured to store routing information of the service routers in the service cluster, including information such as a service router address, an IP address, and a port number.
  • the service cluster routing unit 307 is configured to route a type of service transmitted between the subnets according to the service cluster routing table 306; for example, forwarding the message of the service type between different subnets.
  • the routing decision unit 301 is configured to determine a type and a destination address of the message, and determine, according to the type of the message and the destination address, which routing protocol is used to forward the message;
  • the service router address included in the service cluster routing table 306 may include the cluster head node of the service cluster and/or the service router address of other cluster member nodes. This depends on the mode of the routing protocol in the service cluster. If all messages need to be transited through the cluster head node, the service cluster routing table only needs to store the service router address, IP address, port number and other information of the cluster head node. If the routing protocol allows direct communication between member nodes in the cluster, the routing information of other cluster member nodes can also be saved.
  • the cluster member node After the service cluster is formed, the cluster member node only knows the routing information of the cluster head node. When it needs to communicate with other cluster member nodes, the cluster head node can be queried to obtain the routing information of the other party, and the routing information obtained by the query can be obtained. It is stored in its own business cluster routing table. When communicating again later, you can no longer query the cluster head node.
  • the service router that is in the global P2P layer and joins a service cluster has the foregoing functional units 301-307.
  • the routing decision unit 301 may be specifically configured according to the purpose of access.
  • the service type and destination service address, the intra-subnet routing unit 303, the global routing unit 305, or the service cluster routing unit 307 are selected to route the destination service to its access service router.
  • the service router in the embodiment of the present invention changes the unit participating in the work according to the role of the service router in the service network. If the service router only acts as a node in the subnet and does not join the global P2P layer, the global routing unit 305 and the global routing table 304 need not be enabled; if the service router does not join the service cluster, the service cluster routing unit 307 and Service cluster routing table 306.
  • the service router in the embodiment of the present invention may access one or more application servers, or may not access any application server, for example, a service router that joins the global P2P layer.
  • the application server of the embodiment of the present invention can be used to provide various application or resource access services, and the application server can also be a requester of other applications.
  • the embodiment of the invention further provides a networking method for a service network.
  • FIG. 3 is an overall flowchart of a networking method according to an embodiment of the present invention.
  • the service network of the embodiment of the present invention includes multiple subnets, and the subnet includes multiple service routers. As shown in FIG. 3, the method includes:
  • the global service routers of the plurality of subnets constitute a global peer to peer network layer for routing at least one type of service message between different subnets;
  • S102 Create a service cluster by using the global peer to peer network layer and the multiple subnets, where the service cluster is used to route messages of the same type of service within the service cluster.
  • the embodiment of the invention divides the service network into multiple subnets according to a physical network or an administrative domain.
  • the network of a city of a telecom operator can be regarded as a local subnet, and all service routers in the city form a subnet.
  • the subnet can also form a local subnet for service routers that access the same transmission network.
  • all service routers on the same local area network LAN, Loca l Area Ne twork
  • All service routers in the metropolitan area network WAN, Met ropol i tan Area Network
  • the local subnet can select a routing protocol according to the size and the service requirement.
  • the embodiment of the present invention does not limit the routing protocol used in the local subnet, and different local subnets can use different routing protocols for the service routers in the subnet.
  • Message routing The service routers in each subnet can exchange routing information with each other according to the routing protocol determined by the subnet, and configure their respective routing tables.
  • the local subnet routing table information of the service router includes the service network address of the service router in the subnet, the service router IP address, the service router port number, and the access service type.
  • the services in each subnet can access the corresponding service router according to the service type, that is, the same type of service is provided by the same service router. If the number of services exceeds the access capability of a single service router, multiple service routers can be used. Provide access to certain types of services.
  • the above service types include: SMS, voice, Internet, etc.
  • Table 1 is an example of the service type. Please refer to Table 1. For example, the voting interactive service and the SMS query weather service are jointly attributed to the short message service. If there are multiple service routers responsible for the same type of service access in a subnet, one of these routers can join the global P2P layer. If the service router joins the global P2P layer, the service router needs to run two routing protocols, namely the intra-subnet routing protocol and the global routing protocol, which are used for message routing and inter-subnet message routing in the subnet. Table 1
  • the service network can notify each subnet to elect the service router to join the global P2P layer through a management node.
  • the management node can be a service directory server or a DNS (Doma in Naming System) server, or a separate server for managing functions.
  • the management server is used to complete functions such as management of the service network, global P2P initialization, configuration, registration of service clusters, and monitoring of service network conditions.
  • global P2P initialization you can manually configure global P2P parameters and subnet parameters, such as multicast address or broadcast address of each subnet, global P2P protocol, port, or global P2P layer boot node.
  • Each subnet can be configured with an election policy or manually configured to join the service router of the global P2P layer.
  • the candidate nodes can simultaneously run the intra-subnet routing protocol and the global P2P routing protocol.
  • the service routers that are elected to join the global P2P layer first use the intra-subnet routing protocol to query whether the service routers that have joined the global P2P layer are already in the subnet. If so, the service routers that have joined the global P2P layer are used as their bootstrap nodes. Start a new node join process, such as building your own routing table.
  • the advertisement message is sent to the subnet, and the advertisement message includes the service router address and the information that has been added to the global P2P, so that the service router in the subnet can forward the subnet through the global P2P layer node. Message between.
  • the service type information is stored on the service directory server, and the boot node information of the global P2P layer is saved to provide a query service. If a service router that has joined the global P2P layer in a subnet fails due to overload, quit, power outage, etc., other service routers in the subnet will detect the failure of the global P2P layer service router, and thus, monitoring The service router to the failed condition triggers the process of electing a new global P2P layer service router in the subnet, and broadcasts the election message to the subnet.
  • the other service routers in the subnet determine whether to participate in the election according to the preset policy.
  • the default policy is as follows: Whether the load status of the subnet is allowed, whether the service router of the same type has been added to the global P2P layer in the subnet.
  • the service router that initiates the election (such as the service router that detects the failure condition mentioned above) returns a response, otherwise it does not respond. For example, a service router or a lightly loaded service router that is not joined to a specific service cluster, and these candidate service routers return a response to the service router that initiated the election.
  • the service router that initiates the election selects one or more service routers to join the global P2P layer, and returns an election success message to the selected node.
  • the selected service router joins the global P2P layer, and the process is similar to the new node join procedure described above.
  • the service router in the subnet actively requests to join the global P2P layer, it first sends a join request to the service router that has joined the global P2P layer in the subnet, and processes the joining process of the node according to the protocol used by the global P2P layer.
  • the global P2P layer creation process and subsequent maintenance can use structured P2P protocols, such as chord, CAN (Content Addres sable Networks), Pas ry, etc., and the present invention is not limited to what specific protocol.
  • the common suffix of the service address of the service to which it is accessed is used.
  • P2P layer ID As a parameter of its calculation service router P2P layer ID.
  • the routing process of the service message in the global P2P layer also uses this common suffix to calculate the key value of the service. In this way, the message can be finally routed to the service router having the same P2P layer ID as the key value, and the service router having the P2P layer ID value is the service path responsible for the service access of the type.
  • the P2P layer ID of the service router generated by the P2P protocol and the key value of the service may be specified as follows:
  • P2P layer ID of the service router hash (service type. subnet ID. service network ID)
  • Service key hash (service type. subnet ID. service network ID)
  • only one service router that is responsible for accessing the same service in the same subnet can be added to the global P2P layer, so that the uniqueness of the P2P layer ID of the service router can be ensured.
  • the key of the same type of service in the same subnet because the same service address suffix is used to calculate the key value, the same key value is generated.
  • the service is routed to any service router that provides the service access service of the service type.
  • the service router determines, according to the routing decision unit, that the message belongs to the local subnet. For the service, the local routing protocol can be used to forward the message to the final access service router. Therefore, it does not confuse access information between similar services.
  • the service cluster creation process is triggered, that is, the service cluster is initiated. .
  • the service cluster initiating node first queries the service directory in the service network whether the service cluster already exists.
  • the service cluster information can be pre-stored in the service directory of the service network.
  • the type of the service cluster can be pre-configured or defined.
  • the service network can only create a defined service cluster. After the service cluster is created, it needs to be registered in the service directory, mark the service cluster as the active state and save the cluster head node information of the service cluster.
  • the service cluster initiating node determines that the state of a certain type of service cluster is already existing according to the query result, the creation process is terminated; if the type of service is not defined, the service cluster creation process is also terminated. Otherwise, when the status of the service cluster is non-existent and the type of service is defined, the service cluster initiating node creates a notification for broadcasting the service cluster of the service in each subnet through the global P2P layer. After receiving the service cluster creation notification, the service routers that join the global P2P layer in each subnet broadcast the service cluster creation notification to all service routers in the subnet by using the intra-subnet routing protocol.
  • the service router determines whether it belongs to the service class, and if so, returns a service cluster creation response to the service cluster initiating node, where the response includes the description information of the service router, for example, at least one of the following parameters: performance information of the service router , load information, IP address, port number, etc.
  • the service cluster initiating node After collecting the service router information of all the service clusters, the service cluster initiating node selects a service router as the cluster head node of the service cluster according to the preset policy, and uses the global P2P layer to join at least one of each subnet.
  • the service router of the global P2P layer sends a service cluster generation advertisement, where the advertisement includes information of the cluster head node of the service cluster, for example, including at least one of the following: an identifier of the service cluster, a service network address of the cluster head node, and an IP of the cluster head node. Address, port number of cluster head node, service cluster routing protocol, etc.
  • a service router that joins the global P2P layer in each subnet broadcasts the service cluster to generate an advertisement.
  • the service routers that join the service clusters in each subnet send a cluster member registration message to the cluster head node, and the cluster member registration message includes its own service address, IP address, port number, and access service. Information such as scope.
  • the cluster head node After collecting the information of each cluster member, the cluster head node saves the cluster member information of the service cluster in the service cluster routing table to form a service cluster.
  • the cluster head node is used to record routing information of other member nodes in the cluster, and to forward messages between member nodes in the cluster or provide services for querying routing information of cluster member nodes.
  • the access service range information included in the cluster member registration message refers to which services the service router provides for access services, such as a short message service or a multimedia service, or a part of a certain type of service, such as a certain number of short message services. Specific business.
  • the embodiment of the present invention may also set conditions for destroying the service cluster, for example, the traffic of a certain service cluster is less than a preset threshold or the number of nodes in the service cluster is less than a preset number, and when the destruction condition is met, the service cluster is destroyed. This makes it possible to dynamically build and destroy business clusters.
  • the embodiment of the invention further provides a message routing method in a service network.
  • FIG. 4 is an overall flowchart of a message routing method according to an embodiment of the present invention. As shown in Figure 4, the method includes:
  • S20 receives a service interaction message, where the service interaction message includes a destination service address and a destination. business type;
  • S202 Route the service interaction message to the access service router of the destination service by using at least one of a subnet, a global peer-to-peer network layer, and a service cluster according to the destination service address and the destination service type.
  • service interaction messages are forwarded and routed through the service routers it accesses.
  • the access service router uses the routing decision unit to determine whether the service destination address is a service in the local subnet. If the destination service address belongs to the subnet, the routing protocol in the subnet is used to route the service interaction message to the access service router of the destination service. Then, the access service router of the destination service forwards the service interaction message to the application server that processes the destination service.
  • routes between service routers in a subnet may use different networking structures and routing schemes according to their respective characteristics.
  • a simple networking scheme that can be used is a mesh network, that is, each service router can directly communicate with any other service router in the subnet.
  • Each service router maintains routing information for all other service routers in the subnet. Before each communication, the routing table is queried to find the corresponding service router routing address, and then communicates directly.
  • the service router can use the destination service address information included in the service interaction message to route the service interaction message to the access service router of the destination service.
  • FIG. 5 is a flowchart 1 of signaling interaction of a route within a subnet according to an embodiment of the present invention. As shown in Figure 5, the process includes:
  • the S 30 source service application server (for example, the video information application) sends a service interaction message to the access service router 1 of the source service to request access to the destination service (such as a short message voting service); the service interaction message includes a destination service address, for example Vote. sms. Shenzhen, s erv i cene t work, com;
  • the access service router 1 of the source service determines that the destination service belongs to the subnet according to the destination service address; S303.
  • the access service router 1 of the source service routes the service interaction message to the next hop service router 2 according to the local subnet routing protocol and the local routing table (s r63. sms. Shenzhen, servi cen etwork. com); for example
  • the routing table of service router 1 is shown in Table 2:
  • the service router 2 queries the access service list, and determines that the destination service belongs to the service accessed by itself, that is, the service router 2 is the access service router of the destination service;
  • the service router 1 forwards the service interaction message to the destination service application server ( vote. sms. shenzhen. servi cenetwork.com);
  • FIG. 6 is a second flowchart of signaling interaction of intra-subnet routes according to an embodiment of the present invention.
  • the process includes: The service application server (for example, the leg shup service) sends a service interaction message to the access service router 1 of the source service to request access to the destination service (for example, querying the ticket service); the service interaction message includes the destination service address, for example, t icket, sms. Shenzhen, servicenetwo rk. com;
  • the access service router 1 (s r235. mushup. Shenzhen, servicewor k. com) of the source service determines that the destination service belongs to the subnet according to the destination service address;
  • the access service router 1 of the source service routes the service interaction message to the next hop service router 2 according to the local subnet routing protocol and its local routing table (s r63. sms. Shenzhen, servicene twork. com); Column ⁇ , routing table of service router 1 is shown in Table 3:
  • the service router 2 determines that the destination service belongs to the local subnet according to the destination service address, and does not belong to the service scope of the access.
  • the service router 1 routes the service to the next hop service router according to the local subnet routing protocol and the local routing table.
  • the service router 3 determines that the destination service belongs to the service accessed by itself, that is, the service router 3 is the access service router of the destination service;
  • the service router 3 forwards the service interaction message to the destination service application server.
  • the destination service application server returns a response to the source service application server.
  • a service router uses the routing decision unit to determine whether the destination service address belongs to the internal subnet. If it belongs to the subnet, the local subnet routing protocol is used to route the message; if it is not in the subnet, it is determined whether the destination service belongs to the service cluster joined by the service router, and if so, the intra-cluster routing protocol is used to route the message.
  • the message if no, forwards the service interaction message to another service router of the same type as the access service router of the destination service in the subnet, or forwards it to the service router that joins the global P2P layer in the subnet.
  • the foregoing process is specifically: when the service router determines that the destination service type belongs to the service cluster that the user belongs to, the service router queries the cluster head node for the address of the access service router to which the destination service address belongs. The cluster head node queries the intra-cluster routing table and returns the access service route of the destination service. Address. After the service router obtains the address of the access service router of the destination service, the service router forwards the message to the access service router of the destination service.
  • the specific process may be: the cluster member node in the service cluster forwards the service interaction message to the cluster head node; the cluster head node forwards the message to the access service router of the destination service according to the destination service address and the routing information. Business interaction message.
  • the service interaction message received by the service router does not belong to the service cluster that the service router joins, it is forwarded to the service router in the subnet that is responsible for accessing the destination service type. If the service router that is responsible for accessing the destination service type finds that there is no service cluster of the service type, or if there is no access service router of the destination service type in the local subnet, the service is forwarded through the global P2P layer routing protocol. The interactive message is sent to a service router of the destination service home subnet.
  • the service router may cache the service cluster information locally after querying the service directory or the service cluster information of the domain name server. Set the time to save the cache information, and delete the cache information after it expires. The next time the business cluster information is queried, it is queried directly from the local cache, so that it is no longer necessary to query the business directory.
  • FIG. 7 is a flowchart of signaling interaction of message routing in a service cluster according to an embodiment of the present invention. As shown in Figure 7, the process includes:
  • the source service application server (ops tar. sms. Shenzhen, serv i cene twork. com) sends a service interaction message to the access service router 1 (s r542.sms. Shenzhen, servicene twork.com) of the source service;
  • the service router 1 determines that the destination service belongs to a service in a service cluster, that is, a service cluster in which a short message type exists in the service network;
  • the service router 1 queries the cluster head node 2 of the service cluster (s r523. sms. guangzhou. s ervicenetwork.com) for the address of the access service router of the destination service;
  • the cluster head node 2 queries the intra-cluster routing table, and returns the access service router address of the destination service (s r244. sms be ij ing. servicenetwork.com); for example, Table 5 is the header node 2 Intra-cluster routing table;
  • the service router 1 routes the service interaction message to the access service of the destination service based on the service cluster routing protocol in the service cluster according to the access service router address (sr244. sms be ijing. service network) of the destination service. Router 3 (sr244. sms beijing. servicen e twork. co-m);
  • the access service router of the destination service forwards the service interaction message to the destination service application server (news.sms.bei jing.servicenetwork.com);
  • the destination service application server returns a response to the source service application server.
  • FIG. 8 is a flowchart 2 of signaling interaction of message routing in a service cluster according to an embodiment of the present invention. As shown in Figure 8, the process includes:
  • the access service router 1 determines whether the destination service belongs to the local subnet according to the destination service address. If it belongs to the local subnet, it is based on the local subnet in the local subnet.
  • the network routing protocol is configured to route the service interaction request message to the access service router of the destination service. If it is determined that the destination service does not belong to the local subnet, determine whether the destination service belongs to the service in the existing network according to the destination service type.
  • the service router 1 determines that the destination service belongs to a service cluster in the existing network, and then executes S603;
  • the access service router of the source service forwards the service interaction message to the service router 2 (popstar.sms.shenzhen.servicenetwork.com) added to the service cluster in the subnet;
  • the service router 2 determines that the destination service belongs to the service cluster.
  • the service router 2 queries the cluster head node 3 (sr523.sms. guangzhou. s ervicenetwork.com) of the service cluster to query the address of the access service router of the destination service;
  • the cluster head node 3 queries the intra-cluster routing table, and returns the address of the access service router of the destination service (sr244.sms beijing.servicenetwork.com); as shown in Table 6, Table 6 is the cluster of the node 3 Routing table.
  • the service router 2 routes the service interaction message to the access service router 4 of the destination service according to the service cluster routing protocol according to the access service router address of the destination service (s r244. sms be ij Ing. serv icenetwork. com);
  • the access service router 4 of the destination service forwards the service interaction message to the destination service application server (news.sms.be i ing. servicene twork.com);
  • the destination service application server returns a response to the source service application server.
  • the routing decision unit of the service router finds that the destination service address of the service interaction message does not belong to the local subnet, and cannot forward the message through the service cluster, the service interaction message needs to be routed through the routing protocol in the subnet to the subnet.
  • the service router of the global P2P layer is the service router of the global P2P layer.
  • the global service router obtains the information of the destination service, calculates the key value of the destination service according to the foregoing rule, and uses the key value and the global routing table saved by the node to route the service interaction message to the same P2P layer ID value as the key.
  • Business router The routing process of the message may vary according to different P2P algorithms.
  • the key value is gradually routed to the service router having the P2P layer ID close to the key value, and finally routed to the same P2P layer ID as the above key value. Value of the service router.
  • FIG. 9 is a flowchart of signaling interaction of message routing between subnets according to an embodiment of the present invention. As shown in Figure 9, the process includes the following steps:
  • the source service application server (voicema i 1. mul t imedia. shenzhen. servicene t work, com) sends a service interaction message to its access service router 1 (s r221. mul t imed ia, shenzhen. servicenetw ork. com)
  • the access service router 1 of the source service determines that the destination service is not based on the destination service address. It belongs to the local subnet; and, it is determined that the service cluster corresponding to the destination service type does not exist;
  • the access service router 1 of the source service forwards the service interaction message to the service router 2 (s r231. shenzhen. serv icenetwork.com) that has joined the global P2P layer according to the global routing table and the global routing protocol;
  • the service router 2 determines that the destination service belongs to the local subnet according to the destination service address, and routes the service interaction message to the next hop service router 3 in the global P2P layer based on the global routing protocol and the global routing table. ;
  • the service router 3 (s r463. guangzhou. serv icenetwork.com), according to the destination service address, determines that the destination service belongs to the local subnet, and routes the service interaction message to the existing routing protocol and the global routing table.
  • Next hop service router 4 of the global P2P layer (s r573. be ij ing. servi cenetwork. com);
  • the service router 4 determines that the destination service belongs to the local subnet according to the destination service address, and forwards the service interaction message to the access service router 5 of the destination service according to the local subnet protocol and the local routing table;
  • the access service router of the destination service forwards the service interaction message to the destination service application server.
  • the S71 destination service application server returns a response to the source service application server.
  • FIG. 1 is a network architecture diagram of a service network system according to an embodiment of the present invention.
  • the service network in the embodiment of the present invention includes multiple local subnets (Loca l Sub-Network), and each local subnet is composed of multiple service routers.
  • Each subnet includes at least one global P2P layer node, and a global overlay network (Globa l Over lay) is formed between these nodes.
  • Each service router is responsible for accessing certain types of services or applications. All service interaction messages of such services are forwarded and routed through their corresponding service routers.
  • service routers responsible for access services of the same type of services can form service clusters. These service routers responsible for the same type of service access are forwarding In this type of service message, the message can be directly forwarded in the service cluster to speed up the addressing speed of the message, reduce the number of intermediate forwarding, and reduce the response time of the service.
  • the service network adopts a unified addressing scheme according to the network structure, and all service routers in the local subnet are addressed as follows:
  • a service router of the subnet of the business network in Shenzhen which is responsible for the message service, can be addressed as s r001. sms. sz. gd. serv icenetwork. com; where s rOOl represents the service router identifier, Sms indicates that the service type is short message service, s z. gd stands for Shenzhen City, Guangdong province, corresponding to the subnet identifier, and servi cenetwork.com is the service network identifier.
  • the address of a service router that is not responsible for specific service access is s r024. sz. gd. serv i cene twork. com.
  • the present invention does not limit the addressing scheme, and it is possible to use different addressing schemes according to the specific deployment of the operator, such as addressing in a network-level manner.
  • Each local subnet can organize service routers in the local subnet through various routing protocols to ensure that each service router in the subnet can directly or indirectly forward messages to other service routers in the subnet.
  • the local subnet may also have a self-organizing mechanism. For example, the service router can dynamically join and quit.
  • the routing protocol can ensure the accuracy and integrity of the routing information of each service router.
  • the management node can elect a global P2P layer node according to a preset policy, or manually assign a service router to join the global P2P layer.
  • the election algorithm can consider only the bandwidth and CPU parameters.
  • Each service router participating in the election calculates its priority by itself and returns the priority to the election initiation node.
  • the election initiation node determines which service router joins the global P2P layer according to the priority, and gives the selected service router. Send election success message.
  • Service routers that are elected or designated to join the global P2P layer use their own service network address Calculate the P2P layer ID value.
  • the P2P layer ID value of the service router in the global P2P layer uses the common suffix of the service address of its access service as its calculation P2P.
  • the parameter of the layer ID The routing process of the message in the global P2P layer also uses this common suffix to calculate the key value of the service.
  • the service router that is requesting to join the global P2P layer queries the subnet for the existence of a service router that has joined the global P2P layer.
  • the query mode can be performed by sending a broadcast message to the subnet, or by using a device such as a DNS or a service directory, and using the obtained node as a boot node to construct a self-route table.
  • the node is used as the boot node of the node that is requesting to join the global P2P layer. If the node of the global P2P layer is not added to the local subnet, the node that has joined the global P2P layer of the external network is searched through the service directory or DNS.
  • the initialization of the global P2P layer can also be done manually.
  • the first node of the global P2P layer is set, and the information of the global P2P layer is registered to the service directory or DNS, including the list of nodes that have joined the global P2P layer.
  • the routing table data may be obtained according to the routing algorithm used by the global P2P layer, and the other nodes of the global P2P layer are notified to join the event.
  • FIG. 10 is a flowchart of creating a global P2P layer according to an embodiment of the present invention. As shown in FIG. 10, the process includes the following steps:
  • the S80 management node initiates the process of electing a node that joins the global P2P layer, and uses the intra-subnet routing protocol to send a broadcast message in the subnet to elect a global P2P layer node.
  • the service router determines whether the condition for participating in the election is met according to a preset policy.
  • Each service router participating in the election returns its own priority to the management node;
  • one or more service routers with the highest priority are selected to join the global P2P layer, and if there are nodes with the same priority, one may be randomly selected;
  • the management node returns an election result to the service router selected to be added to the global P2P layer, and triggers a process in which the service router starts to join the global P2P layer.
  • the selected service router queries the subnet to check whether there is a service router that has joined the global P2P layer.
  • step S909 is performed; if the response is received, the node that sends the response is used as the boot node, and the new node is started.
  • the subsequent joining process the specific joining process may use the new node joining process in the P2P protocol, for example, constructing a routing table, notifying a subsequent node according to the calculated P2P layer ID and the obtained routing information, and details are not described herein.
  • the service directory is queried to the node that the other subnet has joined the global P2P layer; S810.
  • the service directory returns a list of nodes that have been added to the global P2P layer by other subnets.
  • the service router that is selected to join the global P2P layer uses the global P2P layer node in the received node list as the boot node to start the subsequent joining process of the new node. .
  • the service-related clusters can be dynamically created according to the type of routing messages between subnets.
  • Service routers responsible for the same type of service access form a cluster that forwards messages directly without routing messages through the global P2P layer.
  • a service router or a management node in the global P2P layer in the service network detects that the traffic of a certain type of service meets a predetermined policy on the global P2P layer, the service cluster of the service is started.
  • FIG. 11 is a flowchart of creating a service cluster according to an embodiment of the present invention.
  • the process includes: S90: When any service router or management node (hereinafter referred to as a service cluster initiating node) in the global P2P layer monitors that the message of the class A service satisfies a preset policy, triggering creation of the service cluster .
  • the preset policy includes that the number or proportion of messages exceeds a preset threshold, the number of service routers that are responsible for accessing the service exceeds a predetermined threshold, the average number of hops of the route exceeds a threshold, and the average number of hops of the route is required for the message to reach the destination application server.
  • the number of times of forwarding for example, if the routing message of the short-term voting service needs to be forwarded 6 times to reach the destination application server, the average number of hops is 6 times.
  • the service cluster initiating node queries the service directory whether the service cluster already exists.
  • the service directory returns service cluster information, including information about whether the service cluster is predefined, status, or the like;
  • the service cluster initiating node broadcasts a class A service cluster creation notification through the global P2P layer, for example, a mashup class service cluster creation notification; otherwise, the creation process is terminated;
  • the service router that joins the global P2P layer in each subnet uses the intra-subnet routing protocol to notify the service router responsible for accessing the service class.
  • each service router that receives the service cluster creation notification initiates a node to the service cluster Return information such as performance, load, IP address, port number, etc.
  • the service cluster initiating node After collecting the service router information of all the service clusters, the service cluster initiating node selects a service router as the cluster head of the service cluster according to a preset policy, and the policy of selecting the cluster head may be similar to the election algorithm in the previous section. ;
  • the service cluster initiating node sends the service cluster generation advertisement to each subnet by using a global P2P layer, where the advertisement includes information of a cluster head node of the service cluster;
  • the service router that joins the global P2P layer in each subnet broadcasts the service cluster to the subnet to generate an advertisement.
  • the service routers that join the service clusters in each subnet send a cluster member registration message to the cluster head node, where the message includes its own service address, IP address, port number, and access service range.
  • the cluster head node After collecting the information of each cluster member node, the cluster head node saves the cluster member information of the service cluster in the service cluster routing table to form a service cluster.
  • the service cluster head sends a service cluster acknowledgement message to its service cluster member
  • the service cluster head registers service cluster information with the service directory.
  • Service clusters can reduce the number of route hops and response times for specific types of services, but they also increase the complexity of the network and the overhead of additional network structure maintenance. In order to ensure the manageability and maintainability of the network, the benefits of creating a service cluster are greater than the maintenance cost. It is not suitable to create a large number of service clusters in the service network, but must create service clusters according to requirements.
  • the service cluster can be created according to the current traffic characteristics in the service network, and when the traffic cluster and other characteristics of the service cluster no longer meet the predetermined conditions, the service cluster needs to be destroyed.
  • the service cluster destruction process is relatively simple, and only needs to instruct the service cluster head node to destroy the service cluster, and the cluster head node broadcasts a service cluster destruction notification to all nodes in the cluster.
  • the cluster head updates the cluster head information and updates the status of the service cluster to an inactive state.
  • the embodiment of the present invention divides the service network into a local subnet, a service cluster, and a global P2P layer. Effectively, the service is connected to the service network according to the service, which facilitates the management and control of the service, and also limits most of the local service interaction traffic to the local subnet.
  • the cross-subnet service interaction uses the global P2P layer routing to ensure the self-organization ability of the entire service network, so that the service routers joining the global P2P layer can dynamically join and exit the network without manual intervention, which reduces the network operation. cost.
  • the service cluster can establish a "shortcut" for the same type of service interaction between multiple subnets, reducing the number of route hops for the same type of service interaction, improving the response speed of the service, shortening the service time of the service, reducing the burden on the global P2P layer, and improving the efficiency of service interaction. .
  • the service network networking method and the routing method of the embodiments of the present invention can improve the reliability of the service network, enhance scalability, and reduce network operation costs.
  • the storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Acces s Memory (RAM).
  • ROM Read-Only Memory
  • RAM Random Acces s Memory

Landscapes

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

Description

一种业务路由方法和业务网络 本申请要求于 2010 年 3 月 17 日提交中国专利局、 申请号为 201010127189. 2 , 发明名称为 "一种业务路由方法和业务网络" 的中国专利 申请的优先权， 其全部内容通过引用结合在本申请中。 技术领域 本发明涉及一种业务路由方法和业务网络， 具体涉及业务网络的组网方 法、 消息路由方法、 路由器及业务网络。 背景技术 互联网技术的迅猛发展促进了互联网业务和内容的极大丰富， 使用户对 互联网产生了更强的黏度和忠诚度。 由于互联网业务的廉价性， 特别一些业 务免费提供给用户， 用户通常选择使用互联网业务替代电信业务， 这对电信 运营商产生了极大的冲击。 各运营商逐步由单一的话音运营商向综合信息运 营商进行转型。 在向综合信息运营商的转型过程中， 运营商迫切需要大量的 业务 /信息来支撑其提供的业务的多样性和丰富性， 从而增强用户的忠诚度。 从技术上来说， 运营商必须拥有有效的业务层基础设施来支持新业务的快速 开发、 大量业务和资源 （包括第三方业务和资源） 的有效组织和管理， 从而 最终为用户提供可管理、 有质量保证的服务。
特别地， 随着移动互联网的兴起， 各种多媒体业务层出不穷， 移动宽带 业务已经开始出现爆炸性增长的态势。 运营商需要及时把握这种业务发展趋 势， 积极开发互联网业务与电信能力融合的组合业务。 这就要求运营商具备 对电信和互联网等不同网络领域的业务集成的能力， 使之成为可控、 可运营、 有质量保证的业务。
但是， 随着业务特性的增加和业务复杂度的提高， 目前各类业务提供商 和运营商在开发、 管理业务的过程中使用的平台架构、 技术纷繁复杂， 导致 业务开发緩慢、 成本高昂、 业务互联互通复杂、 管理混乱等问题。 因此， 在 互联网业务与电信业务逐渐融合的过程中， 为了减少整个业务流程体系中的 业务间交互障碍和复杂度， 迫切需要一种技术框架来规范业务与业务之间的 发现、 交互、 运行、 管理、 支撑机制， 从而增强业务之间的交互效率、 降低 运营支撑成本。 发明内容 本发明实施例提供了一种业务网络的组网方法、 消息路由方法、 路由器 及业务网络。
一方面， 本发明实施例提供了一种业务网络的组网方法， 所述业务网络 包括多个子网， 所述子网包括多个业务路由器， 所述方法包括： 从每个所述 子网的多个业务路由器中确定至少一个全局业务路由器， 所述多个子网中的 全局业务路由器组成用于在不同子网间路由至少一种类型业务消息的全局对 等网络层； 通过所述全局对等网络层和所述多个子网创建业务簇， 所述业务 簇用于在业务簇内路由同一种类型业务的消息。
另一方面， 本发明实施例提供了一种业务网络的消息路由方法， 所述方 法包括： 接收业务交互消息， 所述业务交互消息包含目的业务地址和目的业务 类型； 根据所述目的业务地址和目的业务类型， 通过子网、 全局对等网络层、 业务簇中的至少一个， 将所述业务交互消息路由至所述目的业务的接入业务 路由器。
还一方面， 本发明实施例提供了一种路由器， 所述路由器包括： 子网内 路由表， 用于存储子网内的业务路由器的路由信息； 子网路由内路由单元， 用于根据所述子网内路由表， 路由子网内传输的业务。
最后一方面， 本发明实施例还提供了一种业务网络， 所述业务网络包括： 多个子网， 所述子网包括多个业务路由器和至少一个全局业务路； 一个全局 对等网络层， 由每个所述子网中全局业务路由器而形成， 用于在不同子网间 路由至少一种类型业务的消息； 至少一个业务簇， 由多个子网中负责同类业 务接入的业务路由器组成， 用于在业务簇内路由同一种类型业务的消息。
本发明实施例提供的技术方案， 通过将业务网络分为本地子网、 业务簇、 全局对等网络层， 有效地使业务按照业务归属接入业务网络， 便于业务的管 理和控制。 而跨子网的业务交互釆用全局对等网络层路由， 保证了整个业务 网络的自组织能力。 业务簇可以在多个子网间的同类业务交互组建捷径， 减 少同类业务交互的路由跳数， 加快业务响应， 减轻全局对等网络层的负担。 附图说明 为了更清楚地说明本发明实施例或现有技术中的技术方案， 下面将对实 施例或现有技术描述中所需要使用的附图作一简单地介绍， 显而易见地， 下 面描述中的附图仅仅是本发明的一些实施例， 对于本领域普通技术人员来讲， 在不付出创造性劳动性的前提下， 还可以根据这些附图获得其他的附图。
图 1为本发明实施例的业务网络的网络架构示意图；
图 2为本发明实施例的业务路由器的功能框图；
图 3为本发明实施例的组网方法的整体流程图；
图 4为本发明实施例的消息路由方法的整体流程图；
图 5为本发明实施例的子网内路由的信令交互流程图一；
图 6为本发明实施例的子网内路由的信令交互流程图二；
图 7为本发明实施例的业务簇内消息路由的信令交互流程图一； 图 8为本发明实施例的业务簇内消息路由的信令交互流程图二； 图 9为本发明实施例的子网间消息路由的信令交互流程图；
图 10为本发明实施例的全局对等网络层的创建流程图；
图 11为本发明实施例的业务簇的创建流程图。 具体实施方式 为使本发明实施例的目的、 技术方案和优点更加清楚， 下面将结合本发 明实施例中的附图， 对本发明实施例中的技术方案进行清楚、 完整地描述， 显然， 所描述的实施例是本发明一部分实施例， 而不是全部的实施例。 基于 本发明中的实施例， 本领域普通技术人员在没有做出创造性劳动前提下所获 得的所有其他实施例， 都属于本发明保护的范围。
本发明实施例提供的业务网络 ( Serv i ce Ne twork )是在物理的网络层之 上建立的一个虚拟的叠加业务网络， 为所有业务提供统一、 安全、 高效的协同 运行、 支撑环境。
本发明实施例的业务网络的组网方法， 将业务网络划分为多个子网， 每 个子网内可以根据子网的需求和特点釆用不同的路由协议， 使得本地业务由 本地业务路由器提供接入服务。 从每个子网中选举或人工指定至少一个业务 路由器加入到全局对等 （P2P , Peer to Peer ) 网络层（全局 P2P层） ， 以进 行子网之间的消息路由。 同时按照业务类型在多个子网之间动态创建业务类 型相关的业务簇， 以加速该类型业务的路由， 改善用户的体验。 本发明实施例的业务网络包括多个子网、 一个全局对等网络层及至少一 个业务簇。 子网可以按照物理网络或管理域划分， 每个子网中包括多个业务 路由器， 一个业务路由器可作为一个节点。 每个子网内可以按照各自的规模、 业务需求、 管理要求等特征来选择路由协议。 每个子网内可以选举或指定至 少一个业务路由器加入全局 P2P层， 即每个子网还包括至少一个全局业务路 由器， 该全局 P2P层负责子网之间的消息的路由， 即可以用于在不同子网间 路由至少一种类型的业务。 不同子网间的负责同类业务接入的业务路由器可 以组成该类业务簇， 该业务簇通过上述全局 P2P层和所述多个子网创建形成， 同类业务的交互消息可直接由簇内节点进行转发。
可选地， 上述业务簇可以包括： 簇头节点和多个簇成员节点， 簇头节点 用于保存多个簇成员节点的路由信息， 并在这多个簇成员节点之间转发消息； 簇成员节点间通过簇头节点来进行通信， 例如簇头节点用于转发簇成员节点 之间交互的消息。 或簇头节点向各簇成员节点提供查询服务， 以使各簇成员 节点根据查询结果直接转发消息， 而不需要通过簇头节点转发。
以下举例说明本发明实施例的业务网络。 如图 1 所示， 本发明实施例的 作为一个举例的业务网络 10包括：
三个子网 12、 14、 16 , 分别用于负责各自子网内的消息的路由， 每个子 网可以按照各自的规模、 业务需求、 管理要求等特征来选择各自的路由协议； 实际运行中， 子网的个数可以多于三个；
一个全局 P2P层 18 , 从每个子网内可以选举或指定至少一个业务路由器 创建成全局 P2P层， 用于负责子网之间的消息的路由；
业务簇 20, 由上述子网中负责 A类业务接入的业务路由器组成， 用于在 业务簇 20的簇内节点之间转发 A类业务的交互消息；
业务簇 22 , 由上述子网中负责 B类业务接入的业务路由器组成， 用于在 业务簇 22的簇内节点之间转发 B类业务的交互消息。 本发明实施例还提供了一种业务路由器， 如图 2 所示， 本发明实施例的 业务路由器 30包括：
子网内路由表 302 , 用于存储子网内其他业务路由器的路由信息， 例如可 以包括业务路由器的业务网络地址、 IP地址、 端口号等， 子网内路由表的信 息会随着釆用的路由协议不同而不同；
子网内路由单元 303 , 用于根据子网内路由表 302 , 路由子网内传输的业 务， 例如转发目的地址为子网内的消息。 进一步， 子网内路由单元还用于维 护子网内拓朴结构。
可选地， 在本发明实施例中， 处于全局 P2P层的业务路由器还可以进一 步包括：
全局路由表 304 , 用于存储加入全局 P2P层的业务路由器的路由信息， 可 以包括业务路由器的 P2P层标识 ID、 IP地址、 端口号等信息；
全局路由单元 305 , 用于根据全局路由表 304 , 路由子网间传输的至少一 种类型的业务； 例如转发子网间的消息， 并参与维护全局 P2P层的拓朴结构。
路由决策单元 301 , 用于判断消息的类型及目的地址， 并根据消息的类型 及目的地址决策釆用何种路由协议转发该消息；
在本发明实施例中， 业务或资源在业务网络内具有全局唯一的编址， 业 务的编址方案可以由运营商自行设计， 例如可以釆用类似以下编址格式： 业务标识.业务类型.子网标识.业务网络标识。
例如： e_tour. mushup. Shenzhen, serv i cene twork. com
业务路由器在业务网络内也具有全局唯一的编址， 编址方案类似于业务 的编址方案， 例如： 业务路由器标识. [业务类型] .子网标识.业务网络标识， 其中业务类型为可选。
例如： s r0124. mushup. Shenzhen, serv icenetwork. com
业务和接入该业务的业务路由器可具有相同的业务网络地址后缀： 业务 类型.子网标识.业务网络标识。
业务路由器的 P2P层 ID的形式为字符与数字组合形成的字符串， 可计算 获得， 如算法： MD5 (业务路由器的业务网络地址后缀） ， 其中 MD5是一种具 体的散列函数。
可选地， 在本发明实施例中， 加入某类业务簇的业务路由器还可以进一 步包括：
业务簇路由表 306 , 用于存储本业务簇内的业务路由器的路由信息， 包括 业务路由器地址、 IP地址、 端口号等信息；
业务簇路由单元 307 , 用于根据业务簇路由表 306 , 路由子网间传输的一 种类型的业务； 例如在不同的子网间转发本业务类型的消息。
路由决策单元 301 , 用于判断消息的类型及目的地址， 并根据消息的类型 及目的地址决策釆用何种路由协议转发该消息； 需要说明的是， 业务簇路由表 306 中包含的业务路由器地址， 可以包括 业务簇的簇头节点和 /或其它簇成员节点的业务路由器地址。 这取决于业务簇 内路由协议的模式， 如果设定所有消息都需要通过簇头节点中转， 则业务簇 路由表只需要保存簇头节点的业务路由器地址、 IP地址、 端口号等信息。 如 果路由协议允许簇内成员节点之间直接通信， 则还可以保存其他簇成员节点 的路由信息。 由于在业务簇形成后， 簇成员节点只知道簇头节点的路由信息， 当需要与其他簇成员节点通信时， 可以通过查询簇头节点来获得对方的路由 信息， 并可以将查询得到的路由信息保存在自己的业务簇路由表中， 以后再 次通信时， 可以不用再查询簇头节点。
可选地， 在本发明实施例中， 既处于全局 P2P层又加入了某一业务簇的 业务路由器具有上述功能单元 301-307 , 此时路由决策单元 301 , 具体可以用 于根据接入的目的业务类型和目的业务地址， 选择子网内路由单元 303、全局 路由单元 305或业务簇路由单元 307路由所述目的业务至其接入业务路由器。
本发明实施例的业务路由器， 根据自身在业务网络中参与的角色， 参与 工作的单元会发生变化。 如果业务路由器仅作为子网内的节点， 而不加入全 局 P2P层， 则不需要启用全局路由单元 305和全局路由表 304; 如果业务路由 器没有加入业务簇， 则不需要启用业务簇路由单元 307和业务簇路由表 306。
本发明实施例的业务路由器可以接入一个或一个以上的应用服务器， 也 可以不接入任何应用服务器， 例如加入全局 P2P层的业务路由器。
本发明实施例的应用服务器可以用于提供各种应用或资源访问服务， 应 用服务器同时也可以是其他应用的请求者。 本发明实施例还提供了一种业务网络的组网方法。 图 3 为本发明实施例 的组网方法的整体流程图。 本发明实施例的业务网络包括多个子网， 所述子 网包括多个业务路由器， 如图 3所示， 该方法包括：
S10 从每个所述子网的多个业务路由器中确定至少一个全局业务路由 器， 所述多个子网中的全局业务路由器组成用于在不同子网间路由至少一种 类型业务消息的全局对等网络层；
S102、 通过所述全局对等网络层和所述多个子网创建业务簇， 所述业务 簇用于在业务簇内路由同一种类型业务的消息。
以下通过进一步的实施例进行详细说明。
本发明实施例按照物理网络或管理域将业务网络划分为多个子网。 例如 可以将电信运营商一个城市的网络作为一个本地子网， 城市内所有的业务路 由器组成一个子网。 子网还可以将接入到同一个传输网络的业务路由器组成 一个本地子网， 例如将所有在同一个局域网 （LAN, Loca l Area Ne twork ) 的 业务路由器组成一个子网， 或将属于同一个城域网（WAN, Met ropol i tan Area Network )内的所有业务路由器组成一个子网。 本地子网可以根据规模、 业务 需求选择路由协议， 本发明实施例不限定本地子网内釆用的路由协议， 不同 的本地子网可以釆用不同的路由协议用于本子网内业务路由器之间的消息路 由。 每个子网内的业务路由器可以根据所属子网确定的路由协议相互交换路 由信息， 并配置各自的路由表。 业务路由器的本地子网路由表信息包括子网 内业务路由器的业务网络地址、 业务路由器 IP地址、 业务路由器端口号、 接 入业务类型等。
每个子网内的业务可以按照业务类型接入相应的业务路由器， 即同一类 业务由同一个业务路由器提供接入， 如果某一类业务数量超出单个业务路由 器的接入能力， 可由多个业务路由器提供某类业务的接入。 上述的业务类型 包括： 短信类、 语音类、 互联网类等， 表 1为业务类型示例表， 请参阅表 1 , 例如投票互动业务和短信查询天气业务共同归属于短信类业务。 如果一个子 网内存在多个负责同一类业务接入的业务路由器， 这些路由器中有一个可以 加入全局 P2P层。 如果业务路由器加入了全局 P2P层， 则该业务路由器需要 运行两种路由协议， 即子网内路由协议和全局路由协议， 分别用于本子网内 消息路由和子网间消息路由。 表 1
Figure imgf000011_0001
业务网络可以在各本地子网形成后， 通过一个管理节点通知各个子网选 举业务路由器加入全局 P2P层。 管理节点可以是一台业务目录服务器或 DNS ( Doma in Naming Sys tem , 域名服务器）服务器， 也可以是单独设置的一台 完成管理功能的服务器。 该管理服务器用于完成业务网络的管理、 全局 P2P 初始化、 配置、 登记业务簇和监控业务网络状况等功能。 在全局 P2P初始化 前， 可通过人工配置全局 P2P 的参数和各子网参数， 如各子网的组播地址或 广播地址， 全局 P2P协议、 端口， 或全局 P2P层引导节点等。
每个子网可以设定选举策略或人工配置加入全局 P2P层的业务路由器， 候选的节点能够同时运行子网内路由协议及全局 P2P路由协议。 被选举加入 全局 P2P层的业务路由器首先利用子网内路由协议查询子网内是否已有加入 全局 P2P层的业务路由器， 如果有， 则以该已加入全局 P2P层的业务路由器 作为其引导节点来启动新节点加入过程， 例如构建自己的路由表等。 或如果 本子网内没有已加入全局 P2P层的节点， 则通过 DNS或业务目录服务器等设 备获取已加入全局 P2P层的节点列表， 通过该节点列表中的节点作为引导节 点， 生成各自的路由表记录， 并通知相关节点其已加入全局 P2P层。 例如， 在业务路由器加入全局 P2P层后， 向子网内发送通告消息， 该通告消息包括 业务路由器地址、 已加入全局 P2P 的信息， 以便子网内业务路由器可以通过 该全局 P2P层节点转发子网间的消息。
上述业务目录服务器上保存了业务类型信息， 及全局 P2P层的引导节点 信息等， 以提供查询服务。 如果某个子网内的已加入全局 P2P层的业务路由器由于发生过载、 退出、 断电等情况而失效， 该子网内将有其他业务路由器监测到该全局 P2P层业务 路由器的失效， 从而， 监测到失效情况的业务路由器触发子网内选举新的全 局 P2P层业务路由器的过程， 向子网内广播选举消息。
子网内的其他业务路由器根据预设的策略决定是否参与选举， 预设的策 略例如为： 自身的负载情况是否允许， 本子网内是否已经有同类型的业务路 由器加入到了全局 P2P层， 当根据预设的策略判断参与选举时， 向发起选举 的业务路由器 （如前述监测到失效情况的业务路由器）返回响应， 否则不响 应。 例如未加入特定业务簇的业务路由器或负载较轻的业务路由器， 这些候 选业务路由器向发起选举的业务路由器返回响应。 发起选举的业务路由器从 中选择一个或一个以上业务路由器加入全局 P2P层， 并向被选中的节点返回 选举成功消息。 被选中的业务路由器加入全局 P2P层， 过程与上文描述的新 节点加入过程类似。
如果子网内的业务路由器主动申请加入全局 P2P层， 则先向本子网内已 加入全局 P2P层的业务路由器发送加入请求， 并按照该全局 P2P层所釆用的 协议处理节点的加入过程。
全局 P2P 层的创建过程及后续的维护可以釆用结构化 P2P 协议， 例如 chord, CAN ( Content Addres sable Networks , 内容寻址网络) 、 Pas t ry等, 本发明并不限定釆用何种具体的协议。
为了全局 P2P层在路由过程中保证业务能准确地被路由到其接入业务路 由器， 在确定全局 P2P层中的业务路由器的 P2P层 ID时， 釆用其所接入业务 的业务地址的共同后缀作为其计算业务路由器 P2P层 ID的参数。 而业务消息 在全局 P2P层中的路由过程， 也是利用这个共同后缀来计算该业务的 key值。 这样就可以保证该消息最终被路由到具有与该 key值相同的 P2P层 ID的业务 路由器， 而具有该 P2P层 ID值的业务路由器即为负责该类业务接入的业务路 本发明实施例中，可对 P2P协议生成的业务路由器 P2P层 ID和业务的 key 值做如下规定：
业务路由器的 P2P层 ID = hash (业务类型.子网标识.业务网络标识） 业务 key = hash (业务类型.子网标识.业务网络标识）
在本发明实施例中， 可规定同一个子网内负责同类业务接入的业务路由 器最多只有一个可以加入全局 P2P层， 从而可以保证业务路由器的 P2P层 ID 的唯一性。在同一子网内的同一类业务的 key, 由于釆用相同的业务地址后缀 来计算 key值， 所以会生成同样的 key值。 但由于只需要利用该 key值， 将 该业务路由至提供该类业务接入服务的任意一个业务路由器， 该业务路由器 接收到该消息后， 会根据路由决策单元判断该消息属于本地子网内的业务， 则可以釆用本地路由协议转发该消息至最终的接入业务路由器。 因此， 并不 会对同类业务间的访问信息造成混淆。
当全局 P2P层中的任一业务路由器监测到任一类业务的子网间流量超过 了预设的阔值， 或满足其他预设的条件后， 触发业务簇创建过程， 即作为业 务簇发起节点。 在创建该业务簇之前， 该业务簇发起节点先在业务网络中向 业务目录查询是否已经存在该类业务簇。
业务簇信息可以预先保存在业务网络的业务目录中， 业务簇的类型可以 预先配置或定义， 业务网络只能创建已定义的业务簇。 在业务簇创建后， 需 要在业务目录中登记， 标记该业务簇为活动状态及保存该业务簇的簇头节点 信息。
如果该业务簇发起节点根据查询结果判断某类业务簇的状态为已存在， 则终止创建过程； 如果该类型业务未被定义， 也终止业务簇创建过程。 否则， 当该业务簇的状态为不存在， 且该类型业务为已定义， 则该业务簇发起节点 通过全局 P2P层向各子网内广播该类业务的业务簇创建通知。 每个子网内的 加入全局 P2P层的业务路由器在接收到该业务簇创建通知后， 利用子网内路 由协议向所属子网内的所有业务路由器广播该业务簇创建通知。 每个子网内 的业务路由器判断自身是否属于该业务类， 如果属于， 则向业务簇发起节点 返回业务簇创建响应， 该响应中包括该业务路由器的描述信息， 例如至少包 括以下一种参数： 业务路由器的性能信息、 负载信息、 IP地址、 端口号等。
业务簇发起节点在收集了所有加入该业务簇的业务路由器信息后， 根据 预设的策略选择一个业务路由器作为该业务簇的簇头节点， 并利用全局 P2P 层向每个子网的至少一个已加入全局 P2P层的业务路由器发送业务簇生成通 告， 该通告包括该业务簇的簇头节点的信息， 例如包括以下至少一种： 业务 簇的标识、 簇头节点的业务网络地址、 簇头节点的 IP地址、 簇头节点的端口 号、 业务簇路由协议等。 每个子网内加入全局 P2P层的业务路由器向子网内 广播该业务簇生成通告。 各子网内加入该业务类簇的业务路由器收到业务簇 生成通告后， 向簇头节点发送簇成员登记消息， 该簇成员登记消息包括自身 的业务地址、 IP地址、 端口号、 接入业务范围等信息。 簇头节点在收集了各 簇成员的信息后， 保存该业务簇的簇成员信息在业务簇路由表中， 形成业务 簇。
上述簇头节点， 用于记录簇内其他成员节点的路由信息， 并为簇内成员 节点之间转发消息或提供查询簇成员节点路由信息的服务。
上述簇成员登记消息中包含的接入业务范围信息指该业务路由器为哪些 业务提供接入服务， 例如短信类业务或者多媒体业务， 或者是某类业务的一 部分， 比如短信类业务包含的某几个具体的业务。
本发明实施例还可以设定销毁业务簇的条件， 比如某一业务簇流量小于 一个预设的阔值或业务簇内节点数小于预设的数量， 当满足销毁条件时， 销 毁该业务簇， 从而可以实现动态地建立和销毁业务簇。 本发明实施例还提供了一种业务网络中的消息路由方法。 图 4 为本发明 实施例的消息路由方法的整体流程图。 如图 4所示， 该方法包括：
S20 接收业务交互消息， 所述业务交互消息包含目的业务地址和目的 业务类型；
S202、 根据所述目的业务地址和目的业务类型， 通过子网、 全局对等网 络层、 业务簇中的至少一个， 将所述业务交互消息路由至所述目的业务的接 入业务路由器。
以下通过进一步的实施例进行详细说明。
在一个子网内， 业务交互消息通过其接入的业务路由器转发、 路由。 接 入业务路由器接收到业务交互消息后， 利用路由决策单元判断业务目的地址 是否为本子网内的业务。 如果目的业务地址属于子网内， 则釆用本子网内的 路由协议， 将该业务交互消息路由至目的业务的接入业务路由器。 然后， 目的 业务的接入业务路由器再转发该业务交互消息至处理目的业务的应用服务器。
本发明实施例中， 子网内业务路由器之间的路由可以根据各自的特点釆 用不同的组网结构及路由方案。 例如， 可以釆用的一种简单的组网方案为网 状网， 即每个业务路由器都可以直接与子网内其他任意业务路由器通信。 每 个业务路由器都保持子网内其他所有业务路由器的路由信息。 每次通信前， 都查询路由表找到对应的业务路由器路由地址， 然后直接通信。
业务路由器利用业务交互消息中包含的目的业务地址信息， 路由上述业 务交互消息至该目的业务的接入业务路由器即可。
以下举例说明子网内消息路由的具体过程：
图 5为本发明实施例的子网内路由的信令交互流程图一。 如图 5所示， 该流程包括：
S 30 源业务应用服务器 （例如影视资讯短信应用） 向源业务的接入业 务路由器 1 发送业务交互消息， 以请求访问目的业务（如短信投票业务） ； 该业务交互消息中包括目的业务地址, 例如 vote. sms. Shenzhen, s erv i cene t work, com;
S 302、 源业务的接入业务路由器 1 (s r235. mushup. Shenzhen, serv i cewor k. com)根据目的业务地址， 判断目的业务属于本子网内； S303、 源业务的接入业务路由器 1根据本地子网路由协议和本地路由表， 将该业务交互消息路由至下一跳业务路由器 2 (s r63. sms. Shenzhen, servi cen etwork. com)； 例如， 业务路由器 1的路由表如表 2所示：
表 2
Figure imgf000016_0001
5304、业务路由器 2 (s r63. sms. shenzhen. servicenetwork. com)查询自身 接入业务列表， 判断目的业务属于自身接入的业务， 即业务路由器 2 为目的 业务的接入业务路由器；
5305、 业务路由器 1向目的业务应用月良务器（ vote. sms. shenzhen. servi cenetwork. com )转发该业务交互消息；
5306、 目的业务应用服务器向源业务应用服务器（ pops tar. sms. shenzhe n. servicenetwork. com )返回响应。 图 6为本发明实施例的子网内路由的信令交互流程图二。 如图 6所示， 该流程包括： 5401、 源业务应用服务器（例如腿 shup类业务）向源业务的接入业务路 由器 1 发送业务交互消息， 以请求访问目的业务（如查询车票业务） ； 该业 务交互消息中包括目的业务地址, 例如 t icket, sms. Shenzhen, servicenetwo rk. com;
5402、 源业务的接入业务路由器 1 (s r235. mushup. Shenzhen, servicewor k. com)根据目的业务地址， 判断目的业务属于本子网内；
5403、 源业务的接入业务路由器 1 根据本地子网路由协议和自身的本地 路由表， 将该业务交互消息路由至下一跳业务路由器 2 (s r63. sms. Shenzhen, servicene twork. com)； 列 ^口， 业务路由器 1的路由表 ^口表 3所示：
表 3
Figure imgf000017_0001
5404、 业务路由器 2才艮据目的业务地址， 判断目的业务属于本子网内， 且不属于自身接入的业务范围；
5405、 业务路由器 1根据本地子网路由协议和本地路由表， 将该业务交 互 消 息 路 由 至 下 一 跳 业 务 路 由 器
3 (s r234. sms. shenzhen. servicenetwork. com)； 例如, 业务路由器 2 的路由 表 ^口表 4所示：
表 4
Figure imgf000018_0001
5406、 业务路由器 3 (s r234. sms. shenzhen. servicenetwork. com)判断目 的业务属于自身接入的业务， 即业务路由器 3为目的业务的接入业务路由器；
5407、 业务路由器 3向目的业务应用服务器转发该业务交互消息；
5408、 目的业务应用服务器向源业务应用服务器返回响应。 当某业务路由器接收一个业务交互消息后， 利用路由决策单元判断目的 业务地址是否属于本子网内部。 如果属于本子网内， 则釆用本地子网路由协 议路由该消息； 如果不属于本子网内， 则判断目的业务是否属于本业务路由 器加入的业务簇， 如果是， 则利用簇内路由协议路由该消息， 如果否， 则将 该业务交互消息转发至本子网内与目的业务的接入业务路由器同类的其它业 务路由器进行处理， 或者转发至本子网内加入全局 P2P层的业务路由器处理。
上述过程具体为： 当该业务路由器判断接收到目的业务类型属于自身加 入的业务簇时， 该业务路由器向簇头节点查询目的业务地址所属的接入业务 路由器地址。 簇头节点查询簇内路由表， 并返回该目的业务的接入业务路由 器地址。 该业务路由器获得目的业务的接入业务路由器地址后， 转发该消息 至目的业务的接入业务路由器。 上述具体过程还可以为： 业务簇内的簇成员 节点将所述业务交互消息转发给簇头节点； 簇头节点根据目的业务地址和所 述路由信息， 向目的业务的接入业务路由器转发所述业务交互消息。
如果业务路由器接收到的业务交互消息不属于本业务路由器加入的业务 簇， 则转发至本子网内负责接入目的业务类型的业务路由器处理。 如果该负 责接入目的业务类型的业务路由器发现不存在该类业务的业务簇， 或在本地 子网内不存在目的业务类的接入业务路由器时， 则通过全局 P2P层路由协议， 转发该业务交互消息至目的业务归属子网的某个业务路由器。
在业务路由器查询业务簇的过程中， 业务路由器可以在查询业务目录或 域名服务器某类业务簇信息后， 緩存该业务簇信息在本地。 设定保存该緩存 信息的时间， 过期后则删除该緩存信息。 下次查询该业务簇信息时， 直接从 本地緩存中查询， 从而不必再查询业务目录。
以下举例说明本发明实施例业务簇内消息路由的具体过程。
图 7为本发明实施例的业务簇内消息路由的信令交互流程图一 。 如图 7 所示， 该流程包括：
S501、源业务应用服务器 ( ops tar. sms. Shenzhen, serv i cene twork. com ) 向源业务的接入业务路由器 1 ( s r542. sms. Shenzhen, servicene twork. com ) 发送业务交互消息； 上述业务交互消息中包含目的业务地址（news. sms. be i j i ng. servicenetwork. com)和目的业务类型 (type=sms app l ica t ion)；
5502、 业务路由器 1 判断目的业务属于一个业务簇内的业务， 即业务网 络中存在短信类的业务簇；
5503、 业务路由器 1向该业务簇的簇头节点 2 ( s r523. sms. guangzhou. s ervicenetwork. com )查询目的业务的接入业务路由器地址；
5504、 簇头节点 2 查询簇内路由表， 并返回该目的业务的接入业务路由 器地址 (s r244. sms be i j ing. servicenetwork. com)； 例如, 表 5为褒头节点 2 的簇内路由表；
表 5
Figure imgf000020_0001
5505、 业务路由器 1根据目的业务的接入业务路由器地址（sr244. sms be ijing. servicenetwork) , 在所述业务簇内， 基于业务簇路由协议， 将该业务 交互消息路由至目的业务的接入业务路由器 3(sr244. sms beijing. servicen e twork. co-m)；
5506、 目的业务的接入业务路由器 3转发该业务交互消息至目的业务应 用服务器 (news. sms. bei j ing. servicenetwork. com)；
S507、 目的业务应用服务器向源业务应用服务器返回响应。
图 8 为本发明实施例的业务簇内消息路由的信令交互流程图二。 如图 8 所示， 该流程包括：
S601、 源业务应用服务器 (magazine, mul t imedia. Shenzhen, servicenetw ork. com)向源业务的接入业务路由器 1 (sr952. multimedia. Shenzhen, servic enetwork. com)发送业务交互消息； 上述业务交互消息中包含目的业务地址（n ews. sms. bei j ing. servicenetwork. com)和目的业务类型 (type=sms appl ica t ion)；
5602、 源业务的接入业务路由器 1 在接收至上述业务交互消息后， 根据 目的业务地址判断目的业务是否属于本地子网内； 如果属于本地子网内， 则 在本地子网内， 基于本地子网路由协议， 将该业务交互请求消息路由至目的 业务的接入业务路由器； 如果判断目的业务不属于本地子网内， 则根据目的 业务类型判断所述目的业务是否归属于现有网络中的业务簇之一， 本实施例 中，业务路由器 1判断目的业务归属于现有网络中的一个业务簇，则执行 S603;
5603、 源业务的接入业务路由器 1 将该业务交互消息转发至本子网内加 入 该 类 业 务 簇 的 业 务 路 由 器 2 (popstar. sms. shenzhen. servicenetwork.com) ；
5604、 业务路由器 2判断目的业务属于业务簇内；
5605、 业务路由器 2向该业务簇的簇头节点 3 ( sr523. sms. guangzhou. s ervicenetwork. com )查询目的业务的接入业务路由器地址；
5606、 簇头节点 3 查询簇内路由表， 并返回该目的业务的接入业务路由 器地址 (sr244. sms beijing. servicenetwork.com); 如表 6所示, 表 6为褒 头节点 3的簇内路由表。
表 6
Figure imgf000021_0001
Figure imgf000022_0001
S607、 业务路由器 2根据目的业务的接入业务路由器地址， 在所述业务 簇内， 基于业务簇路由协议， 将该业务交互消息路由至目的业务的接入业务 路由器 4 (s r244. sms be i j ing. serv icenetwork. com)；
S608、 目的业务的接入业务路由器 4转发该业务交互消息至目的业务应 用服务器 (news. sms. be i j ing. servicene twork. com)；
S609、 目的业务应用服务器向源业务应用服务器返回响应。 当业务路由器的路由决策单元发现业务交互消息的目的业务地址不属于 本地子网， 并且不能通过业务簇路由消息后， 需要将该业务交互消息通过子 网内的路由协议路由至本子网内已加入全局 P2P层的业务路由器。
该全局业务路由器获取目的业务的信息， 根据前述规则计算目的业务的 key值， 并利用该 key值和本节点保存的全局路由表， 路由该业务交互消息至 与该 key具有相同的 P2P层 ID值的业务路由器。 消息的路由过程根据不同的 P2P算法会有差异， 一个实施例是， 利用 key值逐渐路由到具有与 key值接近 的 P2P层 ID的业务路由器， 最后路由至具有与上述 key值相同的 P2P层 ID 值的业务路由器。
图 9为本发明实施例的子网间消息路由的信令交互流程图。 如图 9所示， 该流程包括如下步骤：
S701、 源业务应用服务器 (voicema i 1. mul t imedia. shenzhen. servicene t work, com)向其接入业务路由器 1 (s r221. mul t imed ia, shenzhen. servicenetw ork. com)发送业务交互消息； 该业务交互消息中包含目的业务地址（vo icema i 1. mul t imedia. be i j ing. serv icene twork. com)和目的业务类型 (type=mul t im edia appl ica t ion)；
S702、 源业务的接入业务路由器 1 根据目的业务地址， 判断目的业务不 属于本地子网； 并且， 判断目的业务类型对应的业务簇不存在；
S703、 源业务的接入业务路由器 1 根据全局路由表和全局路由协议， 转 发该业务交互消息到本地子网内 已加入全局 P2P 层的业务路由器 2 (s r231. shenzhen. serv icenetwork. com)；
S704-S705 , 业务路由器 2才艮据目的业务地址判断目的业务属于本地子网 之外， 则基于全局路由协议和全局路由表， 将业务交互消息路由至处于全局 P2P层的下一跳业务路由器 3;
S706-S707. 业务路由器 3 (s r463. guangzhou. serv icenetwork. com)根据 目的业务地址， 判断目的业务属于本地子网之外， 则基于全局路由协议和全 局路由表， 将业务交互消息路由至处于全局 P2P 层的下一跳业务路由器 4 (s r573. be i j ing. servi cenetwork. com)；
S708-S709 , 业务路由器 4根据目的业务地址， 判断目的业务属于本地子 网内， 则根据本地子网协议和本地路由表， 将业务交互消息转发到目的业务 的接入业务路由器 5 ;
S710、 目的业务的接入业务路由器 5转发该业务交互消息到目的业务应 用服务器；
S71 目的业务应用服务器向源业务应用服务器返回响应。 以下举例进一步详细说明本发明的上述具体实施例。
请再次参阅图 1 , 图 1为本发明实施例的业务网络系统的组网架构图。如 图 1所示，本发明实施例的业务网络包括多个本地子网（Loca l Sub-Network ), 每个本地子网由多个业务路由器组成。 每个子网内包括至少一个全局 P2P层 节点， 这些节点之间形成一个全局的叠加网络（Globa l Over lay)。 每个业务 路由器负责某类业务或应用的接入， 这类业务的所有业务交互消息都通过其 对应的业务路由器进行转发、 路由。 而不同子网内， 负责同类业务接入服务 的业务路由器可以形成业务簇。 这些负责同类业务接入的业务路由器在转发 该类业务消息时， 可以直接在业务簇内转发消息， 以加快消息的寻址速度， 减少中间转发次数， 减少业务的响应时间。
该业务网络按照网络结构釆用统一的编址方案， 本地子网内的所有业务 路由器都按如下方案编址：
业务路由器标识. [业务接入类型. ]子网标识.业务网络标识
例如， 业务网络的广东省深圳市子网的负责消息类业务接入的某个业务 路由器可编址为 s r001. sms. sz. gd. serv icenetwork. com; 其中, s rOOl 表示 业务路由器标识， sms表示业务类型为短信类业务， s z. gd代表广东省深圳市， 对应于子网标识, servi cenetwork. com为业务网络标识。
而某个未负责具体业务接入的业务路由器的编址则为 s r024. sz. gd. serv i cene twork. com。 关于编址， 本发明并不限定编址的方案， 可以才艮据运营商 具体的部署釆用不同的编址方案， 如按照网络层次方式编址。
各本地子网可以通过各种路由协议组织本地子网内的业务路由器， 保证 子网内每个业务路由器均可直接或间接转发消息至子网内其他业务路由器。 本地子网也可能具备自组织机制， 例如业务路由器可以动态的加入、 退出， 路由协议能保证各业务路由器的路由信息的准确性和完整性。
在本地子网构造完成后，管理节点可以根据预先设定的策略选举全局 P2P 层节点， 或者通过人工配置方式指定加入全局 P2P层的业务路由器。 上述预 先设定的策略可以是基于多种选举算法， 例如可以根据业务路由器的性能、 存储容量、 负载或带宽等指标的加权值来衡量候选的业务路由器优先度， 例 如 业 务路 由 器 的 优 先 度 =CPU* 30°/o+memory- s ize*10%+ (l-load) * 30% +bandwidth* 30%o 可选地， 选举算法可以只考虑 bandwidth和 CPU两个参数。 每个参与选举的业务路由器自行计算其优先度， 并将该优先度返回给选举发 起节点， 选举发起节点 （管理节点）根据优先度确定哪个业务路由器加入全 局 P2P层， 并给选定的业务路由器发送选举成功消息。
被选举或指定加入全局 P2P层的业务路由器利用自己的业务网络地址计 算 P2P层 ID值。 为了全局 P2P层在路由过程中保证业务能准确地路由到其接 入业务路由器， 在全局 P2P层中的业务路由器的 P2P层 ID值釆用其接入业务 的业务地址的共同后缀作为其计算 P2P层 ID的参数。 而消息在全局 P2P层中 路由过程也是利用这个共同后缀计算该业务的 key值。
假设在业务网络的广东深圳子网内有一个负责短信类业务接入的业务路 由 器 ， 其 P2P 层 ID 值 计 算 如 下 ： P2P 层 ID=hash (sms. sz. gd. servicene twork. com)。 通过这种方式, 就可以保证该消 息最终被路由到与其 key值相同的 P2P层 ID的业务路由器， 而具有该 P2P层 ID值的业务路由器就是负责该类业务接入的业务路由器。
正在请求加入全局 P2P层的业务路由器在子网内查询是否存在已加入全 局 P2P层的业务路由器。 该查询方式可以通过向子网内发送广播消息， 或通 过 DNS、 业务目录等设备来查询， 并将查询得到的节点作为引导节点， 以构建 自已的路由表。
如果本地子网内已有节点加入全局 P2P层， 则以该节点作为正在请求加 入全局 P2P层的节点的引导节点。 如果本地子网内没有加入全局 P2P层的节 点， 则通过业务目录或 DNS查找外网的已加入全局 P2P层的节点。
全局 P2P层的初始化也可以通过人工设置的方式完成， 设置全局 P2P层 的第一个节点， 并将全局 P2P层的信息登记到业务目录或 DNS , 包括已加入全 局 P2P层的节点列表等信息。 在新节点获得引导节点信息后， 可以根据全局 P2P层釆用的路由算法获取路由表数据，并通知全局 P2P层其他节点该新节点 加入事件等。
本地子网内除初始化阶段需要选举业务路由器加入全局 P2P层外， 在本 地子网内已加入全局 P2P层的业务路由器失效后， 需要选举替代节点加入全 局 P2P层时也会触发全局 P2P层的业务路由器选举过程。 子网内也可以有业 务路由器主动加入全局 P2P层， 其过程和上述过程类似。 图 10为本发明实施例的全局 P2P层的创建流程图。 如图 10所示， 该流 程包括以下步骤：
S80 管理节点发起选举加入全局 P2P层的节点的过程， 利用子网内路 由协议发送子网内广播消息， 以选举全局 P2P层节点；
5802、 业务路由器接收到选举消息后， 根据预设的策略判断是否满足参 与选举的条件； 预设的策略包括子网内是否有负责同类接入服务的业务路由 器已加入全局 P2P层、 是否满足最低性能要求、 是否超过负荷上限等； 如果 满足参与选举条件， 则可进一步根据下式计算本节点的优先度： 优先度 =CPU* 30% + memory. s i ze* 10% + (1-l oad) * 30% + bandwi dth* 30%;
5803、 各参与选举的业务路由器向管理节点返回自身的优先度；
5804、 管理节点接收到所有参与选举的节点返回的优先度信息后， 选择 最高优先度的一个或多个业务路由器加入至全局 P2P层， 如果有相同的优先 度的节点， 可以随机选择一个；
5805、 管理节点向被选中加入全局 P2P层的业务路由器返回选举结果， 触发这些业务路由器启动加入全局 P2P层的流程；
5806、 被选中的业务路由器计算其 P2P层 ID, 例如可以根据下式计算： P2P层 ID=ha sh (接入业务类型.子网标识.业务网络标识）；
5807、 被选中的业务路由器向子网内查询是否有已加入全局 P2P层的业 务路由器；
5808、 等待预定的时间， 如果没有接收到响应， 则表明子网内没有加入 全局 P2P层的节点， 则执行步骤 S909 ; 如果接收到响应， 则利用该发送响应 的节点作为引导节点， 启动新节点的后续加入过程， 具体加入过程可釆用 P2P 协议中新节点加入过程， 例如根据计算的 P2P层 ID和获取的路由信息构建路 由表、 通知后继结点等， 在此不赘述。
5809、 如果被选中的业务路由器判断等待超时， 即在预设的时间内没有 在子网内接收到响应， 则向业务目录查询其他子网已加入全局 P2P层的节点； S810、 业务目录返回其他子网已加入全局 P2P层的节点列表； S81 被选中加入全局 P2P层的业务路由器利用接收到的节点列表中的 全局 P2P层节点作为引导节点， 启动新节点的后续加入过程。 在业务网络中为了加快业务的路由速度及响应时间， 减轻全局 P2P层路 由负载， 可以根据子网间路由消息的类型动态地创建业务相关的簇。 负责同 一类业务接入的业务路由器形成一个簇， 直接转发消息， 而不需要通过全局 P2P层路由消息。当业务网络中某个处于全局 P2P层的业务路由器或管理节点 监测到在全局 P2P层上， 某类业务的流量等特征满足预定的策略， 则启动该 类业务的业务簇的创建。
图 11为本发明实施例的业务簇创建的流程图。 如图 11所示， 该流程包括： S90 当全局 P2P层内任一业务路由器或管理节点 （这里简称业务簇发 起节点）监测到 A类业务的消息满足预设的策略， 触发创建该类业务簇。 预 设的策略包括消息数量或比例超过预设的阔值、 负责接入该类业务的业务路 由器数量超过预定的阈值、 平均路由跳数超过阈值等； 平均路由跳数就是消 息到达目的应用服务器需转发的次数， 例如， 短息投票业务的路由消息若平 均需要经过 6次转发才能到达目的应用服务器， 则平均路由跳数为 6次。
5902、 当 A类业务簇触发条件满足后， 业务簇发起节点向业务目录查询 是否已经存在该类业务簇；
5903、 业务目录返回业务簇信息， 包括该业务簇是否预定义、 状态等信息；
5904、 如果该业务簇不存在， 则业务簇发起节点通过全局 P2P层广播一 个 A类业务簇创建通知， 例如 mashup类业务簇创建通知； 否则， 终止创建过 程；
5905、 每个子网内加入全局 P2P层的业务路由器接收到该业务簇创建消 息后， 利用子网内路由协议通知负责该类业务类接入的业务路由器；
S906-S907 , 每个接收到业务簇创建通知的业务路由器向业务簇发起节点 返回自身的信息， 如性能、 负载、 IP地址、 端口号等；
5908、 业务簇发起节点在收集了所有加入该业务簇的业务路由器信息后， 根据预设的策略选择一个业务路由器作为该业务簇的簇头， 选择簇头的策略 可以和上节中选举算法类似；
5909、 业务簇发起节点利用全局 P2P层向每个子网发送该业务簇生成通 告， 该通告包括该业务簇的簇头节点的信息；
S99、 每个子网内加入全局 P2P层的业务路由器向子网内广播该业务簇生 成通告；
5911、 各子网内加入该类业务簇的业务路由器收到业务簇生成通告后， 向簇头节点发送簇成员登记消息， 该消息包括自身的业务地址、 IP地址、 端 口号、 接入业务范围等信息；
5912、 簇头节点在收集了各簇成员节点的信息后， 保存该业务簇的簇成 员信息在业务簇路由表中， 形成业务簇；
591 3、 业务簇头向其业务簇成员发送业务簇确认消息；
5914、 业务簇头向业务目录登记业务簇信息。
业务簇可以降低特定类业务的路由跳数及响应时间， 但也会增加网络的 复杂性及额外的网络结构维护开销。 为了保证网络的可管理性、 可维护性， 使创建业务簇带来的效益大于维护成本， 在业务网络中不适宜创建大量的业 务簇， 而必须是根据需求来创建业务簇。
业务簇的创建可以根据业务网络中当前的流量特征来创建特定业务簇， 而当该业务簇的流量等特征不再满足预定的条件时， 需要将该业务簇销毁。 业务簇销毁过程相对简单， 只需要指示业务簇簇头节点销毁该业务簇， 簇头 节点向簇内所有节点广播业务簇销毁通知。 簇头更新簇头信息， 将该业务簇 的状态更新为非活动状态。
本发明实施例的技术方案带来的有益效果包括：
本发明实施例通过将业务网络分为本地子网、 业务簇、 全局 P2P层， 有 效地使业务按照业务归属接入业务网络， 便于业务的管理和控制， 同时也使 大部分的本地业务交互流量限制在本地子网内部。 而跨子网的业务交互釆用 全局 P2P层路由， 保证了整个业务网络的自组织能力， 使加入全局 P2P层的 业务路由器可以动态地加入、 退出网络， 不需要人工干预， 降低了网络的营 运成本。 业务簇可以在多个子网间的同类业务交互组建 "捷径" ， 减少同类 业务交互的路由跳数， 提高业务的响应速度， 缩短业务的寻址时间， 减轻全 局 P2P层的负担、 提高业务交互效率。
本发明实施例的业务网络组网方法及路由方法， 可以提高业务网络的可 靠性、 增强可扩展性并降低网络运营成本。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流 程， 是可以通过计算机程序来指令相关的硬件来完成， 所述的程序可存储于 一计算机可读取存储介质中， 该程序在执行时， 可包括如上述各方法的实施 例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体（ Read-Only Memory, ROM )或随机存储记忆体 ( Random Acces s Memory, RAM )等。 参照前述实施例对本发明实施例进行了详细的说明， 本领域的普通技术人员 应当理解： 其依然可以对前述各实施例所记载的技术方案进行修改， 或者对 其中部分技术特征进行等同替换； 而这些修改或者替换， 并不使相应技术方 案的本质脱离本发明实施例各实施例技术方案的精神和范围。

Claims

权 利 要 求
1、一种业务网络的组网方法， 其特征在于， 所述业务网络包括多个子网， 所述子网包括多个业务路由器， 所述方法包括：
从每个所述子网的多个业务路由器中确定至少一个全局业务路由器， 所 述多个子网中的全局业务路由器组成用于在不同子网间路由至少一种类型业 务消息的全局对等网络层；
通过所述全局对等网络层和所述多个子网创建业务簇， 所述业务簇用于 在业务簇内路由同一种类型业务的消息。
2、 根据权利要求 1所述的方法， 其特征在于， 所述子网根据物理网络或 管理域划分。
3、 根据权利要求 1或 2所述的方法， 其特征在于， 一个所述业务路由器 接入一种类型的业务或一部分所述类型的业务， 所述业务路由器与其所接入 的业务具有相同的地址后缀。
4、 根据权利要求 1或 2所述的方法， 其特征在于， 所述确定至少一个全 局业务路由器具体包括：
从每个子网中选举至少一个全局业务路由器， 或根据人工配置在每个子 网中指定至少一个全局业务路由器。
5、 根据权利要求 1所述的方法， 其特征在于， 所述通过所述全局对等网 络层和所述多个子网创建业务簇具体包括：
处于全局对等网络层的发起节点通过全局对等网络层， 向各子网内广播 业务簇创建通知， 所述业务簇创建通知包括待创建的业务簇的业务类型； 当各子网内的业务路由器判断自身属于所述业务类型时， 向所述发起节 点返回业务簇创建响应， 以申请加入所述业务簇；
所述发起节点根据所述业务簇创建响应， 确定所述业务簇的簇头节点； 所述发起节点向各子网内已加入所述业务簇的业务路由器， 发送包含簇 头节点信息的业务簇生成通告； 各子网内已加入所述业务簇的业务路由器， 向所述簇头节点返回簇成员 登记消息；
簇头节点根据所述簇成员登记消息生成业务簇路由表。
6、 根据权利要求 5所述的方法， 其特征在于， 所述业务簇创建响应包括 以下至少一种参数： 业务路由器的性能信息、 负载信息、 地址信息。
7、 根据权利要求 5所述的方法， 其特征在于， 所述业务簇生成通告包括 以下至少一种： 业务簇的标识、 簇头节点的业务网络地址、 簇头节点的 IP地 址、 簇头节点的端口号、 业务簇路由协议。
8、 根据权利要求 1所述的方法， 其特征在于， 每个全局业务路由器用其 地址后缀作为散列函数的参数计算该业务路由器在全局对等网络层的标识。
9、 根据权利要求 1所述的方法， 其特征在于， 通过所述全局对等网络层 和所述多个子网创建业务簇之前还包括： 根据预设的业务簇创建触发策略， 触发业务簇的创建；
所述预设的业务簇创建触发策略包括以下至少一种：
待监测业务类型的业务流量占全局对等网络层流量的比例超过预设阔 值；
待监测业务类型的业务流量在全局对等网络层中超过预设阔值； 待监测业务类型的业务的接入业务路由器数量超过预设阔值；
待监测业务类型的业务从源应用服务器至目的应用服务器的平均路由跳 数超过预设阔值。
10、 根据权利要求 1 所述的方法， 其特征在于， 通过所述全局对等网络 层和所述多个子网创建业务簇之前还包括：
将预先定义的可用于创建业务簇的业务类型， 以及已存在的业务簇的状 态信息保存于业务目录或域名服务器中。
11、 根据权利要求 10所述的方法， 其特征在于， 通过所述全局对等网络 层和所述多个子网创建业务簇之前还包括： 所述处于全局对等网络层的发起节点通过查询业务目录或域名服务器来 判断待创建的业务簇是否已经存在， 以及判断待创建的业务簇的业务类型是 否为预先定义的业务类型；
当待创建的业务簇不存在， 且待创建的业务簇的业务类型为预先定义的业 务类型时， 所述处于全局对等网络层的发起节点执行所述业务簇的创建过程。
12、 一种业务网络的消息路由方法， 其特征在于， 所述方法包括： 接收业务交互消息， 所述业务交互消息包含目的业务地址和目的业务类型； 根据所述目的业务地址和目的业务类型， 通过子网、 全局对等网络层、 业务簇中的至少一个， 将所述业务交互消息路由至所述目的业务的接入业务 路由器。
1 3、 根据权利要求 12所述的方法， 其特征在于， 所述根据所述目的业务 地址和目的业务类型， 通过子网、 全局对等网络层、 业务簇中的至少一个， 将所述业务交互消息路由至所述目的业务的接入业务路由器具体包括：
当所述目的业务地址属于子网内时， 则通过子网， 基于子网路由协议， 将所述业务交互消息路由至所述子网内目的业务的接入业务路由器； 或者 当所述目的业务地址不属于子网内， 且所述目的业务的类型与业务簇的 类型相同时， 则通过所述业务簇， 并基于业务簇路由协议， 将所述业务交互 消息路由至所述业务簇内目的业务的接入业务路由器； 或者
当所述目的业务地址不属于子网内， 且所述目的业务的类型与业务簇的类 型不同时， 则通过全局对等网络层， 基于全局路由协议， 将所述业务交互消息 路由到目的业务的归属子网， 所述目的业务的归属子网根据子网路由协议， 将 所述业务交互消息转发至所述目的业务的接入业务路由器。
14、 如权利要求 1 3所述方法， 其特征在于， 所述通过业务簇将所述业务 交互消息路由至所述目的业务的接入业务路由器之前还包括：
业务路由器向业务目录或域名服务器查询目的业务是否属于所述目的业 务类型对应的业务簇， 并保存获取的查询结果； 在预设的时间阔值过期后， 删除所述查询结果。
15、 根据权利要求 1 3所述方法， 其特征在于， 所述业务簇包括簇头节点 和多个簇成员节点， 所述簇头节点用于保存所述多个簇成员节点的路由信息； 所述通过业务簇将所述业务交互消息路由至所述目的业务的接入业务路 由器具体包括：
业务簇内的簇成员节点根据目的业务地址， 向簇头节点查询目的业务的 接入业务路由器地址；
簇头节点向所述簇成员节点返回目的业务的接入业务路由器地址； 所述簇成员节点根据目的业务的接入业务路由器地址， 向目的业务的接 入业务路由器转发所述业务交互消息。
16、 根据权利要求 1 3所述方法， 其特征在于， 所述业务簇包括簇头节点 和多个簇成员节点， 所述簇头节点用于保存所述多个簇成员节点的路由信息； 所述通过业务簇将所述业务交互消息路由至所述目的业务的接入业务路 由器具体包括：
业务簇内的簇成员节点将所述业务交互消息转发给簇头节点；
簇头节点根据目的业务地址和所述路由信息， 向目的业务的接入业务路 由器转发所述业务交互消息。
17、 一种路由器， 其特征在于， 所述路由器包括：
子网内路由表， 用于存储子网内的业务路由器的路由信息；
子网内路由单元， 用于根据所述子网内路由表， 路由子网内传输的业务。
18、 根据权利要求 17的路由器， 其特征在于， 所述路由器进一步包括： 全局路由表， 用于存储加入全局对等网络层的业务路由器的路由信息； 全局路由单元， 用于根据所述全局路由表， 路由子网间传输的至少一种 类型的业务；
路由决策单元， 用于根据接入的目的业务类型和目的业务地址， 选择所 述子网内路由单元或全局路由单元路由所述目的业务至其接入业务路由器。
19、 根据权利要求 17的路由器， 其特征在于， 所述路由器进一步包括： 业务簇路由表， 用于存储加入业务簇的业务路由器的路由信息； 业务簇路由单元， 用于根据所述业务簇路由表， 路由子网间传输的一种 类型的业务；
路由决策单元， 用于根据接入的目的业务类型和目的业务地址， 选择所 述子网内路由单元或业务簇路由单元路由所述目的业务至其接入业务路由 哭口
20、 一种业务网络， 其特征在于， 所述业务网络包括：
多个子网， 所述子网包括多个业务路由器和至少一个全局业务路由器； 一个全局对等网络层， 由每个所述子网中全局业务路由器而形成， 用于 在不同子网间路由至少一种类型业务的消息；
至少一个业务簇， 由多个子网中负责同类业务接入的业务路由器组成， 用于在业务簇内路由同一种类型业务的消息。
21、 根据权利要求 20所述的业务网络， 其特征在于， 所述业务簇包括： 簇头节点， 所述簇头节点用于保存多个簇成员结点的路由信息， 并在所 述多个簇成员节点之间转发消息。
PCT/CN2010/080478 2010-03-17 2010-12-30 一种业务路由方法和业务网络 WO2011113289A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP20100847758 EP2541848B1 (en) 2010-03-17 2010-12-30 Service routing method and service network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010127189.2 2010-03-17
CN201010127189.2A CN102195855B (zh) 2010-03-17 2010-03-17 一种业务路由方法和业务网络

Publications (1)

Publication Number Publication Date
WO2011113289A1 true WO2011113289A1 (zh) 2011-09-22

Family

ID=44603271

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/080478 WO2011113289A1 (zh) 2010-03-17 2010-12-30 一种业务路由方法和业务网络

Country Status (3)

Country Link
EP (1) EP2541848B1 (zh)
CN (1) CN102195855B (zh)
WO (1) WO2011113289A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019220447A1 (en) 2018-05-17 2019-11-21 Neragon Networks Ltd Mobile ad-hoc wireless networks
CN112383390A (zh) * 2020-11-12 2021-02-19 刘中亚 一种适用于多播通信的wsn组密钥管理方法
CN113676477A (zh) * 2021-08-20 2021-11-19 中邮科通信技术股份有限公司 基于星状网络和内存路由表实现Socket会话集群通信的方法
WO2022218194A1 (zh) * 2021-04-13 2022-10-20 中国移动通信有限公司研究院 服务路由方法及设备

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9826025B2 (en) * 2013-05-21 2017-11-21 Cisco Technology, Inc. Chaining service zones by way of route re-origination
CN104322019B (zh) * 2013-05-23 2017-11-07 华为技术有限公司 业务路由系统、设备和方法
CN104243302B (zh) 2013-06-20 2018-03-16 华为技术有限公司 业务路由报文处理方法、装置及网络系统
CN103955984B (zh) * 2014-03-24 2017-11-07 广州华多网络科技有限公司 实现互联网投票的方法、服务器、投票处理节点及系统
CN103945487A (zh) * 2014-05-08 2014-07-23 国家电网公司 一种电力通信网中面向可靠性的节点分簇方法
US9729439B2 (en) 2014-09-26 2017-08-08 128 Technology, Inc. Network packet flow controller
CN104468604A (zh) * 2014-12-19 2015-03-25 北京奇虎科技有限公司 局域网中基于对等网络通信模式的数据访问方法及装置
US9736184B2 (en) 2015-03-17 2017-08-15 128 Technology, Inc. Apparatus and method for using certificate data to route data
US9729682B2 (en) 2015-05-18 2017-08-08 128 Technology, Inc. Network device and method for processing a session using a packet signature
CN109660442B (zh) * 2015-09-28 2021-04-27 杭州数梦工场科技有限公司 Overlay网络中组播复制的方法及装置
US9985883B2 (en) 2016-02-26 2018-05-29 128 Technology, Inc. Name-based routing system and method
CN109474905B (zh) * 2018-12-25 2021-08-24 重庆邮电大学 一种基于角色层级化的无线传感网安全路由方法
CN109767106A (zh) * 2018-12-29 2019-05-17 中国联合网络通信集团有限公司 跨组织流程处理方法及装置
CN110545487B (zh) * 2019-08-20 2021-11-16 中央电视台 组播信号编址方法、传输方法及装置、交换机
CN110557336A (zh) * 2019-09-17 2019-12-10 中国人民解放军战略支援部队信息工程大学 一种寻址路由方法及系统
CN115225416B (zh) * 2021-04-19 2024-05-17 北京神州数码云科信息技术有限公司 广播消息的方法、装置、电子设备及存储介质
CN114866559A (zh) * 2022-05-10 2022-08-05 杭州顶天网络有限公司 一种多层次的点对点网络生成方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7246148B1 (en) * 1995-09-29 2007-07-17 Cisco Technology, Inc. Enhanced network services using a subnetwork of communicating processors
CN101136861A (zh) * 2006-09-01 2008-03-05 阿尔卡特朗讯 提供iptv服务的方法
US20080219237A1 (en) * 2007-03-07 2008-09-11 Pascal Thubert Multicast support by mobile routers in a mobile ad hoc network
CN101652959A (zh) * 2007-03-12 2010-02-17 Lm爱立信电话有限公司 涉及网络管理的设备及方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2411806A1 (en) * 2001-11-16 2003-05-16 Telecommunications Research Laboratory Wide-area content-based routing architecture
CN1870624A (zh) * 2005-07-11 2006-11-29 华为技术有限公司 网络中对等实体之间实现业务请求的方法及装置
CN101047638A (zh) * 2006-06-28 2007-10-03 华为技术有限公司 Mesh网路由方法和装置
CN100466859C (zh) * 2006-06-29 2009-03-04 华为技术有限公司 用于无线mesh网的业务QoS保证方法和装置
US8737261B2 (en) * 2006-11-27 2014-05-27 Telefonaktiebolaget L M Ericsson (Publ) Node registering method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7246148B1 (en) * 1995-09-29 2007-07-17 Cisco Technology, Inc. Enhanced network services using a subnetwork of communicating processors
CN101136861A (zh) * 2006-09-01 2008-03-05 阿尔卡特朗讯 提供iptv服务的方法
US20080219237A1 (en) * 2007-03-07 2008-09-11 Pascal Thubert Multicast support by mobile routers in a mobile ad hoc network
CN101652959A (zh) * 2007-03-12 2010-02-17 Lm爱立信电话有限公司 涉及网络管理的设备及方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2541848A4 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019220447A1 (en) 2018-05-17 2019-11-21 Neragon Networks Ltd Mobile ad-hoc wireless networks
EP3794779A4 (en) * 2018-05-17 2022-03-30 Neragon Networks Ltd AD HOC MOBILE WIRELESS NETWORKS
US11523325B2 (en) 2018-05-17 2022-12-06 Neragon Networks Ltd. Mobile ad-hoc wireless networks
US11849307B2 (en) 2018-05-17 2023-12-19 Neragon Networks Ltd. Mobile ad-hoc wireless networks
CN112383390A (zh) * 2020-11-12 2021-02-19 刘中亚 一种适用于多播通信的wsn组密钥管理方法
CN112383390B (zh) * 2020-11-12 2023-06-27 刘中亚 一种适用于多播通信的wsn组密钥管理方法
WO2022218194A1 (zh) * 2021-04-13 2022-10-20 中国移动通信有限公司研究院 服务路由方法及设备
CN113676477A (zh) * 2021-08-20 2021-11-19 中邮科通信技术股份有限公司 基于星状网络和内存路由表实现Socket会话集群通信的方法

Also Published As

Publication number Publication date
EP2541848B1 (en) 2015-05-06
CN102195855B (zh) 2014-10-08
EP2541848A1 (en) 2013-01-02
EP2541848A4 (en) 2013-01-23
CN102195855A (zh) 2011-09-21

Similar Documents

Publication Publication Date Title
WO2011113289A1 (zh) 一种业务路由方法和业务网络
USRE49943E1 (en) System and method for a context layer switch
EP2230802B1 (en) A method and apparatus for maintaining route information
WO2023280166A1 (zh) 跨区域通信方法及设备、计算机可读存储介质
US10050824B2 (en) Managing a cluster of switches using multiple controllers
JP5323861B2 (ja) ネットワーク・リソースをプールするための方法および装置
US20130041982A1 (en) Method and node for acquiring content and content network
US7978631B1 (en) Method and apparatus for encoding and mapping of virtual addresses for clusters
US20090177772A1 (en) Method, system and device for establishing a peer to peer connection in a p2p network
JP2010504668A (ja) リソース配信の方法、システム、およびエッジサーバ
CN112398902B (zh) 高可用负载均衡方法、系统及计算机可读存储介质
WO2010127618A1 (zh) 一种实现流媒体内容服务的系统和方法
WO2022083464A1 (zh) 数据处理方法、装置及系统，计算机设备和计算机可读存储介质
WO2012065531A1 (zh) 实现中继选择的方法及装置、系统
US20120185487A1 (en) Method, device and system for publication and acquisition of content
WO2010069261A1 (zh) 边界网关协议路由处理方法、装置和系统
CN106790502B (zh) 一种基于NAT64前缀的IPv4终端、IPv6服务互通业务的负载均衡系统
Azgin et al. On-demand mobility support with anchor chains in Information Centric Networks
WO2013075636A1 (zh) 分布式哈希表互通网络系统、域间节点及实现方法
WO2013023479A1 (zh) 负荷分担方法、装置和系统
WO2009086763A1 (zh) 一种源切换的方法、系统和设备
WO2013189130A1 (zh) 一种基于点对点网络的通讯系统及通讯方法
WO2013170729A1 (zh) 一种虚拟网络布局的实现方法和系统
Ma et al. Source routing over protocol-oblivious forwarding for named data networking
WO2012175006A1 (zh) 基于点对点的网络管理方法及代理选择服务器

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: 10847758

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2010847758

Country of ref document: EP