WO2021007963A1 - 路由分发方法及控制器、信息路由方法及网络节点设备 - Google Patents

路由分发方法及控制器、信息路由方法及网络节点设备 Download PDF

Info

Publication number
WO2021007963A1
WO2021007963A1 PCT/CN2019/110770 CN2019110770W WO2021007963A1 WO 2021007963 A1 WO2021007963 A1 WO 2021007963A1 CN 2019110770 W CN2019110770 W CN 2019110770W WO 2021007963 A1 WO2021007963 A1 WO 2021007963A1
Authority
WO
WIPO (PCT)
Prior art keywords
client
network
network node
routing
node device
Prior art date
Application number
PCT/CN2019/110770
Other languages
English (en)
French (fr)
Inventor
王力鹏
Original Assignee
厦门网宿有限公司
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 厦门网宿有限公司 filed Critical 厦门网宿有限公司
Priority to EP19937704.5A priority Critical patent/EP3993328A1/en
Priority to US17/627,696 priority patent/US20220329514A1/en
Publication of WO2021007963A1 publication Critical patent/WO2021007963A1/zh

Links

Images

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
    • H04L45/64—Routing or path finding of packets in data switching networks using an overlay routing layer
    • 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
    • H04L45/12—Shortest path evaluation
    • H04L45/124—Shortest path evaluation using a combination of metrics
    • 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
    • H04L45/02—Topology update or discovery
    • 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
    • H04L45/302—Route determination based on requested QoS
    • 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
    • H04L45/302—Route determination based on requested QoS
    • H04L45/306—Route determination based on the nature of the carried application
    • H04L45/3065—Route determination based on the nature of the carried application for real time traffic
    • 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
    • H04L45/42—Centralised routing
    • 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
    • H04L45/66—Layer 2 routing, e.g. in Ethernet based MAN's
    • 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
    • H04L45/74—Address processing for routing

Definitions

  • the present invention relates to the field of Internet technology, in particular to a routing distribution method and controller, an information routing method and network node equipment.
  • the purpose of this application is to provide a route distribution method and controller, information routing method, and network node equipment, which can simplify network deployment and improve network work efficiency.
  • one aspect of this application provides a route distribution method, which is applied to a controller, and the controller stores node detection information sent by each network node device and user detection information sent by each client.
  • the method includes: responding to a first access request from a first client to a second client; based on the node detection information and the user detection information, planning between the first client and the second client Optimal path, and generate a routing strategy corresponding to the optimal path; and distribute the routing strategy to network node devices and/or clients located in the optimal path.
  • another aspect of the present application also provides a controller that stores node detection information sent by each network node device and user detection information sent by each client, and the controller includes: access receiving Unit for receiving the first access request from the first client to the second client; the route planning unit, for generating information between the first client and the second client based on the node detection information and the user detection information The optimal path is planned between the terminals, and the routing strategy corresponding to the optimal path is generated; the route distribution unit is configured to distribute the routing strategy to network node devices and/or clients located in the optimal path.
  • the present application provides an information routing method, which is applied to a network node device.
  • the network node device stores multiple routing strategies issued by a controller, and the routing strategy includes Customer identification
  • the method includes: receiving a data message sent by an upstream node, and identifying a target customer identification carried in the data message, and querying a target routing strategy with the target customer identification; extracting the data message The destination address of the message, and determine the downstream node to which the destination address points from the target routing policy; and forward the data message to the downstream node.
  • the present application also provides a network node device on the other hand.
  • the network node device stores multiple routing strategies issued by the controller, the routing strategy has a customer identifier, and the network node device It includes: a target user identification unit for receiving a data message sent by an upstream node, identifying the target user information carried in the data message, and querying the target routing strategy with the target customer identification; a node management unit, It is used to extract the destination address of the data message, and to determine the downstream node to which the destination address points from the target routing policy; a message forwarding unit is used to forward the data message to the downstream node.
  • the controller calculates the optimal path for data message transmission based on the path decision algorithm based on the node detection information reported by the network node device, and generates the corresponding routing strategy, and then the controller will route The strategy is distributed to each network node device on the optimal path.
  • the network node device only needs to forward the data message according to the received routing strategy, no longer need to care about its own routing table and other traditional entries, and realize the entire network control.
  • the separation of the plane and the data plane simplifies the deployment of the network, and through the adjustment of the path decision algorithm, the technical solution provided by this application can also flexibly adjust the optimal path of data transmission according to the user's actual business needs, and improve the work efficiency of the network .
  • Figure 1 is a schematic diagram of an SDN network architecture in an embodiment of the present invention.
  • Figure 2 is a flowchart of a route distribution method in an embodiment of the present invention.
  • Figure 3 is a schematic diagram of a network topology in an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of functional modules of the controller in the embodiment of the present invention.
  • FIG. 5 is a schematic diagram of the structure of the controller in the embodiment of the present invention.
  • Fig. 6 is a flowchart of an information routing method in an embodiment of the present invention.
  • the traditional network is a distributed control architecture.
  • Each node device (router, switch, etc.) in the network contains an independent control plane and data plane.
  • the control plane used for protocol calculation and the data plane for packet forwarding are located on the same node In the equipment, when the routing calculation and network topology change, each node device has to perform the routing calculation process again, which makes the traditional network unable to flexibly adjust the data transmission path, and the network protocol between each node device is complex, The mechanisms are different, resulting in complicated operation and maintenance when deploying and updating network services.
  • SDN Software Defined Network, Software Defined Network
  • the original distributed control network architecture is reconstructed into a centralized control network architecture.
  • SDN separates the control plane on the node equipment, and the centralized control
  • the controller management does not need to rely on the underlying node equipment, shielding the differences from the underlying node equipment, and the control is completely open, and users can customize any network routing and transmission rules and strategies they want to achieve.
  • Figure 1 is a schematic diagram of the SDN network architecture.
  • the network architecture in SDN is divided into three layers: collaborative application layer, control layer and forwarding layer.
  • the collaborative application layer is mainly used to manage and run various applications developed by users; the control layer is the control center of the entire system.
  • the control layer interacts through the southbound control interface and the forwarding layer, and interacts through the northbound business interface and the collaborative application layer.
  • the main body is the controller.
  • the controller is responsible for the generation of internal switching paths and border service routes of the network, and is responsible for processing network state change events.
  • the controller will Adjust the network switching path and service routing according to the change information of the network status, so that the network is always in a normal working state; the forwarding layer forms a basic forwarding network based on various network node devices in the traditional network, and the forwarding layer reports to the control layer on the one hand Network resource information and status. On the other hand, it receives the routing policy issued by the control layer.
  • the main body of its implementation is the SDN repeater (that is, various network node devices that support SDN).
  • the SDN repeater is responsible for forwarding and forwarding data packets.
  • the forwarding entries required by the process are generated by the control layer, and the SDN repeater is the execution unit of the system and does not make decisions itself.
  • SDN can realize the programmable control of network node equipment through the software platform in the controller, and realize the on-demand deployment of network resources.
  • the SDN repeater is only responsible for pure data message forwarding.
  • the controller has the functions of link discovery and topology management.
  • the link discovery and topology management are mainly used by the controller to uniformly monitor and count the information reported by the underlying network node devices using the uplink channel of the southbound control interface.
  • controller mentioned below has created and stored a complete network topology diagram, and each client and each network node device mentioned below exist in the above network topology diagram.
  • the client, each network node device and the controller can be connected.
  • the route distribution method provided by the present application is applied to a controller, and the controller stores node detection information sent by each network node device and user detection information sent by each client.
  • any network node device in the network can send a detection request message to other network node devices connected to it, and obtain the network node device and the foregoing detection response message according to the detection response message returned by the other network node device. Node detection information between other network node devices.
  • the network node device may use the uplink channel of the southbound control interface to send the node detection information to the controller, so that the controller stores the node detection information. It needs to be pointed out that the network node device can actively send detection request messages regularly according to the preset time value, or passively send detection request messages according to the detection information update instructions issued by the controller when the network topology changes. This application does not restrict this.
  • any client in the network can also send a detection request message to the network node device connected to it, and obtain the client and the network node device according to the detection response message returned by the network node device.
  • the client After obtaining the user detection information, the client can use the uplink channel of the southbound control interface to send the user detection information to the controller, so that the controller stores the user detection information. It needs to be pointed out that the client can actively send detection request messages regularly according to the preset time value, or it can passively send detection request messages according to the detection information update instruction issued by the controller when the network topology changes. , This application does not restrict this.
  • a certain client in the network may be connected to multiple network node devices at the same time.
  • the multiple network node devices mentioned above may be in the same local area network as the client, or may be in a different local area network from the client.
  • the client can only send detection request messages to the network node device in the same local area network as it, so as to reduce the number of detection request messages sent by the client in the network, thereby reducing network load.
  • the client can also only send a detection request message to a network node device that is close to its geographic location and under the same operator's network, which is not limited in this application.
  • node detection information and user detection information at least include network quality information such as network delay, network jitter, and network packet loss rate.
  • network quality information such as network delay, network jitter, and network packet loss rate.
  • the content contained in the node detection information and the user detection information can be more or less, which is not limited in this application.
  • the controller can store the node detection information sent by all network node devices and the user detection information sent by all clients. According to the stored network topology map, the controller can obtain information between all node devices in the network topology map. Network quality information.
  • each client when the route distribution method in this application is applied to the enterprise wide area network, each client also has a customer identification.
  • the above customer identification is bound to the enterprise wide area network identification, and the above customer identification can be used to indicate a certain customer Whether the client is in the same enterprise WAN as other clients.
  • the client when a certain client sends user detection information to the controller, the client also needs to send its client identification to the controller.
  • the route distribution method provided in this application may include the following steps.
  • S101 Receive an access request from a first client to a second client.
  • the first client needs to access the second client.
  • the access request sent by the first client with the destination address of the second client will first arrive at the controller.
  • the above access request carries the first The address of the client and the address of the second client, so that the controller determines the source address and the destination address of the access request according to the address of the first client and the address of the second client.
  • the first client and the second client when the first client and the second client are located in the same enterprise wide area network, the first client and the second client will also have the same customer identifier, and the access request will also carry With the aforementioned customer identification, the first client and the second client with the same customer identification may be located in the same enterprise wide area network.
  • the controller when the controller receives an access request from the first client to the second client, it can first determine whether the corresponding routing policy currently exists. If not, it indicates that the controller has received the access request for the first time. At this time, the first request can be responded to according to the following; if there is, it indicates that the controller has not received the access request for the first time, and at this time, it can forward according to the existing routing strategy. At the same time, the number of requests for access requests can be recorded to confirm the hotspot access requests by statistics, and the corresponding routing strategy can be updated in real time for the hotspot access requests.
  • S102 In response to the first access request from the first client to the second client, based on the node detection information and the user detection information, plan an optimal path between the first client and the second client , And generate the routing strategy corresponding to the optimal path.
  • the controller can use the shortest path algorithm to calculate the optimal path between the first client and the second client according to the stored node detection information and user detection information, and the optimal path includes There are all network node devices through which a data message is sent from the first client to the second client and the upstream and downstream connections between each network node device.
  • the controller generates a corresponding routing strategy according to the optimal path, and the routing strategy includes at least user information and node information.
  • the node information includes at least the address information of all network node devices that a data message passes through when sent from the first client to the second client.
  • the above-mentioned user information includes at least the client identifiers of the first client and the second client, and address information used to characterize the start and destination of the optimal path.
  • the above data message can finally reach the second client according to the above routing strategy.
  • S103 Distribute the routing strategy to network node devices and/or clients located in the optimal path.
  • the controller After the controller generates the routing strategy corresponding to the optimal path, it can use the downstream channel of the southbound control interface to send the routing strategy to all network node devices and/or clients in the optimal path.
  • the controller can uniformly distribute the complete routing strategy to all network node devices and/or clients in the optimal path to reduce control Calculation load of the device.
  • the routing strategy it receives includes both the routing instructions that the network node device or the client needs to execute, and the optimal path described above. Routing instructions to be executed by other network node devices and clients.
  • the controller may further divide the above-mentioned routing strategy into routing sub-items according to the identification of the network node device and the identification of the client in the above-mentioned optimal path.
  • the routing sub-policy is distributed to the corresponding network node device or client.
  • the routing sub-policy only contains the network node device or the client that needs to be executed In order to reduce the network load when the controller issues routing policies, and improve the efficiency of network node equipment or clients when forwarding data messages.
  • the above-mentioned network node device identification and client identification can use the MAC address of the device, or can be determined by a unique number assigned to the device by the controller according to the network topology diagram, which is not limited in this application.
  • the controller can obtain all the connection paths between the first client and the second client (ie, candidate path set) according to the network topology.
  • the connection path includes all network node devices that the data message passes through when sent from the first client to the second client.
  • the controller can identify two directly connected to each other in the stored node detection information according to all the network node devices included in the candidate path and the upstream and downstream connection relationships between each network node device.
  • the node detection information between the network node devices, and the user detection information between the first client and the network node device directly connected to the first client is identified in the stored user detection information, and the second client and the Based on the user detection information between the network node devices directly connected to the second client, the controller can calculate the path index value corresponding to the candidate path based on the node detection information and the user detection information.
  • the controller can calculate the path index values corresponding to all candidate paths in the candidate path set, and determine the optimal path based on the above path index values.
  • the controller When the controller calculates the path index value corresponding to a certain candidate path, it can obtain the network quality parameter corresponding to the candidate path based on the node detection information and user detection information, and according to the above-mentioned network quality parameter and the weight bound to the customer identification.
  • the value parameter is used to calculate the path index value of the candidate path.
  • the above weight parameter can be preset in the controller. In practical applications, because different users may have different business requirements, different weight parameters can be set for different users, and the above weight parameters are bound to the customer ID to distinguish the weights of different users parameter.
  • the above network quality parameters include at least delay parameters and/or jitter parameters and/or packet loss rate
  • the above weight parameters include at least delay weight and/or jitter weight and/or packet loss rate weight. Value, the above weight parameter can be set according to the user’s business type.
  • an enterprise WAN may have multiple branch LANs at the same time, and the terminal devices in each branch LAN do not have public IP addresses, but have different internal network segments, so that the terminal devices in a branch LAN can
  • the above-mentioned client may be a customer premises equipment (CPE), which is used to connect the local area network and the external network. It is important to point out that at this time, a private network can be established between the above-mentioned client and the network node device directly connected to it.
  • CPE customer premises equipment
  • the above-mentioned private network can be based on a public network link and use VPN (Virtual Private Network) to realize the customer
  • VPN Virtual Private Network
  • the information exchange between the client and the network node equipment; the above-mentioned private network can also use the two-layer dedicated line technology to establish a private line connection between the client and the network node equipment, and then realize the information interaction between the client and the network node equipment.
  • the application is not restricted.
  • the client terminal when the client terminal receives a data message sent by the initiating terminal device in a certain local area network and directed to the target terminal device in another local area network, the client can send the data message of the initiating terminal device
  • the internal network segment and the internal network segment of the target terminal device are sent to the controller so that the controller can plan the routing path based on the internal network segment of the originating terminal device and the internal network segment of the target terminal device.
  • the following example illustrates an implementation of the route distribution method provided in this application.
  • CPE1 and CPE2 have the same customer identification 500100
  • CPE3 and CPE4 have the same customer identification 500125
  • network node equipment 1 denoted as POP1
  • network node equipment 2 denoted as POP2
  • Device N (denoted as POPN) is in the same local area network as CPE2, and CPE1, CPE2, CPE3, CPE4, POP1, POP2, and POPN are all located in the network topology diagram of the controller.
  • the controller can determine the candidate paths between CPE1 and CPE2 as CPE1-POP1-POPN-CPE2 and CPE1-POP2-POPN-CPE2 according to the network topology map and the customer identification 500100.
  • the controller Since the controller stores node detection information sent by each network node device and user detection information sent by each client, the controller must store node detection information between POP1 and POP2 (including the delay between POP1 and POP2).
  • Parameter D1 jitter parameter S1 and packet loss rate L1
  • node detection information between POP1 and POPN including delay parameter D2 between POP1 and POPN, jitter parameter S2 and packet loss rate L2
  • POP2 and POPN Node detection information (including the delay parameter D3 between POP2 and POPN, jitter parameter S3 and packet loss rate L3)
  • user detection information between CPE1 and POP1 including the delay parameter D4 between CPE1 and POP1, jitter Parameter S4 and packet loss rate L4
  • user detection information between CPE1 and POP2 including the delay parameter D5 between CPE1 and POP2, jitter parameter S5 and packet loss rate L5)
  • user detection between CPE2 and POPN Information including the delay parameter D6 between CPE2 and POPN,
  • the controller uses the path decision algorithm to calculate each candidate according to the preset weight parameters bound to the customer ID 500100 (assuming the delay weight is ⁇ , the jitter weight is ⁇ , and the packet loss rate weight is ⁇ ).
  • the path index value of the path where:
  • In1 is less than In2, it means that the CPE1-POP1-POPN-CPE2 path is better than the CPE1-POP2-POPN-CPE2 path, and the controller will select the CPE1-POP1-POPN-CPE2 path as the optimal path and generate CPE1-POP1-POPN -The routing strategy corresponding to the CPE2 path, and distribute the generated routing strategy to CPE1, POP1, POPN, and CPE2.
  • the routing strategy is composed of routing instructions to be executed by CPE1, POP1, POPN, and CPE2.
  • the controller can uniformly distribute the complete routing strategy to CPE1, POP1, POPN and CPE2. It can also divide the above routing strategy into routing sub-strategies based on the respective MAC addresses of CPE1, POP1, POPN and CPE2. According to the MAC address, the routing sub-strategies are sent to CPE1, POP1, POPN, and CPE2 respectively, and the routing sub-strategies only include routing instructions that need to be executed by the network node device or the client.
  • POP1 in the CPE1-POP1-POPN-CPE2 path As an example to illustrate how the controller divides and distributes routing sub-strategies. Since data messages in the network can flow in both directions, the routing sub-strategy received by POP1 contains routing instructions in two directions.
  • the controller can query the routing instructions to be executed by POP1 in the routing strategy mentioned above.
  • the routing instructions executed include (destination address CPE1, next hop node CPE1); (destination address CPE2, next hop node POPN), and then the controller only needs to send the above routing instructions to POP1, so that POP1 receives the routing instructions It only contains routing instructions that it needs to execute.
  • the above-mentioned delay weight ⁇ , jitter weight ⁇ , and packet loss rate weight ⁇ can be set according to the actual business needs of the user.
  • a user with a customer ID of 500100 has a delay-sensitive service.
  • the value of the delay weight ⁇ can be increased accordingly;
  • a user with a customer ID of 500125 has a service that is jitter-sensitive
  • the value of the jitter weight ⁇ can be increased accordingly.
  • path decision algorithm listed above is only an example, and is not a limitation on the path decision algorithm. Based on the idea of this application, those skilled in the art can also use other path decision algorithms.
  • the same network node device can be divided into multiple network isolation spaces through network isolation technology, and resource isolation can be formed between each network isolation space, and each network isolation space can be rented by different enterprise users.
  • the network isolation space can be bound to the customer identity.
  • the same network node device can provide data forwarding services to multiple enterprise users at the same time, and distinguish different enterprise users located in the same network node device through customer identification, which can ensure that the routing strategies of different enterprise users do not affect each other.
  • the utilization rate of the system is improved.
  • the node detection information obtained by the network node device can also be reused by different enterprise users, thereby reducing the number of detection request messages sent by the network node device in the network and reducing the network load.
  • the present application also provides a controller that stores node detection information sent by each network node device and user detection information sent by each client, and the controller includes:
  • the access receiving unit is configured to receive an access request from the first client to the second client, determine whether the access request is the first access request, and send the first access request to the routing planning unit;
  • a route planning unit configured to plan an optimal path between the first client and the second client based on the node detection information and the user detection information, and generate a route corresponding to the optimal path Strategies to respond to the first access request received;
  • the routing distribution unit is configured to distribute the routing strategy to the network node devices and/or clients located in the optimal path.
  • the foregoing routing planning unit further includes:
  • a sub-routing planning module configured to divide the routing strategy into routing sub-strategies according to the identification of the network node device in the optimal path and the identification of the client, and distribute the routing sub-strategy to the corresponding network nodes Device or client.
  • the present application also provides a controller, the controller includes a memory and a processor, the memory is used to store a computer program, when the computer program is executed by the processor, it can be implemented as described above Route distribution method.
  • the topic distribution device may include a processor, an internal bus, and a memory.
  • the memory may include memory and non-volatile memory.
  • the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it.
  • the topic distribution apparatus may also include more or less components than those shown in FIG.
  • GPU Graphics Processing Unit
  • external communication ports etc.
  • this application does not exclude other implementations, such as logic devices or a combination of software and hardware.
  • the processor may include a central processing unit (CPU) or a graphics processing unit (GPU), of course, may also include other single-chip microcomputers, logic gate circuits, integrated circuits, etc. with logic processing capabilities, or appropriate combination.
  • the memory described in this embodiment may be a memory device for storing information.
  • the device that can store binary data can be a memory; in an integrated circuit, a circuit with storage function that has no physical form can also be a memory, such as RAM, FIFO, etc.; in the system, it has physical storage
  • the device can also be called a memory, etc.
  • the memory can also be implemented in the form of cloud storage, and the specific implementation manner is not limited in this specification.
  • this application also provides an information routing method, the method is applied to a network node device, the network node device stores a plurality of routing strategies issued by the controller, the routing strategy has a customer Logo.
  • the network node device can serve multiple enterprise WANs at the same time. Therefore, the network node device stores multiple routing policies issued by the controller, and the multiple routing policies have different customer identifications.
  • the above-mentioned customer identification is bound to the enterprise wide area network and the client in the enterprise wide area network, and can be used to determine the routing strategy to be executed by the clients with different customer identifications.
  • the above-mentioned customer identification can also be used to indicate that the source client of the data message is connected with Whether the destination client is in the same enterprise WAN.
  • the above routing strategy is generated by the controller according to the optimal path between the source client and the destination client of the data message.
  • the above routing strategy can be a complete routing strategy, or it can be based on the network node.
  • the routing sub-policy generated by the device ID.
  • the method may include the following steps:
  • S201 Receive a data message sent by an upstream node, identify a target customer identifier carried in the data message, and query a target routing strategy with the target customer identifier.
  • the network node device receives the data message forwarded by the upstream node, and the above data message carries the target client identifier.
  • the above target client identifier may be a client possessed by the source client of the data message.
  • logo The network node device recognizes the above-mentioned target customer identifier, and according to the above-mentioned target customer identifier, searches for the target routing strategy with the above-mentioned target customer identifier among the stored routing strategies, and the above-mentioned target routing strategy contains the corresponding next-hop address
  • the downstream node can be used to forward the data message carrying the above-mentioned target client identifier to the downstream node.
  • S202 Extract the destination address of the data message, and determine the downstream node to which the destination address points from the target routing policy.
  • the aforementioned data message also carries a destination address.
  • the aforementioned destination address may be the address of the destination client of the data message, and the network node device can extract the aforementioned destination address from the aforementioned data message.
  • the next hop address and the downstream node corresponding to the next hop address are determined according to the above-mentioned destination address.
  • the CPE1-POP1-POPN-CPE2 path in the above text illustrates how the network node device determines the next hop address according to the direction of the destination address and the downstream node corresponding to the next hop address. Since data messages in the network can flow in both directions, when forwarding data messages, network node devices first need to determine the flow direction of the received data messages. The flow direction of the above data messages can be based on the data message carried The destination address is determined.
  • the destination address carried in the above data message is the address of CPE2, so POP1 can determine that the flow direction of the above data message points to CPE2, and then POP1 According to the stored routing strategy, it is determined that the next hop address is the address of POPN, and the corresponding downstream node is POPN.
  • S203 Forward the data message to the downstream node.
  • the network node device will forward the above data message to the downstream node corresponding to the next hop address according to the next hop address.
  • a network isolation technology can also be used to divide a plurality of network isolation spaces in the network node device, and the aforementioned network isolation spaces can be bound to a customer identity.
  • the namespace technology can be used to divide multiple network isolation spaces in the network node device, and this application does not limit the technology for achieving network isolation.
  • the network node device can send the routing strategy to the network isolation space with the same customer identifier according to different customer identifiers possessed by the routing strategy.
  • the network node device can be reused by clients with different customer IDs (that is, different enterprise users), which improves the utilization of the system. Accordingly, the network node The node detection information obtained by the device can also be reused by different enterprise users, thereby reducing the number of detection request messages sent by the network node device in the network and reducing the network load.
  • the network node device may also send a detection request message to other network node devices connected to it, and obtain the network node device and the other network node device according to the detection response message returned by the other network node device. Node detection information between devices. After obtaining the node detection information, the network node device can use the uplink channel of the southbound control interface to send the node detection information to the controller, so that the controller stores the node detection information. It needs to be pointed out that the network node device can actively send detection request messages regularly according to the preset time value, or passively send detection request messages according to the detection information update instructions issued by the controller when the network topology changes. This application does not restrict this.
  • a dedicated network can also be established between the network node device and the client.
  • the above-mentioned private network can be based on the public network link, and use VPN (Virtual Private Network) to realize the information interaction between the client and the network node equipment; the above-mentioned private network can also use the two-layer private line technology, and the client and the network A dedicated line connection is established between the node devices to realize the information exchange between the client and the network node device, which is not limited in this application.
  • This application also provides a network node device, which stores multiple routing strategies issued by a controller, the routing strategy has a customer identifier, and the network node device includes:
  • the target user identification unit is configured to receive a data message sent by an upstream node, identify the target user information carried in the data message, and query the target routing strategy with the target customer identifier;
  • a node management unit configured to extract the destination address of the data message, and determine the downstream node to which the destination address points from the target routing policy
  • the message forwarding unit is configured to forward the data message to the downstream node.
  • the aforementioned network node device further includes:
  • a detection information reporting unit configured to obtain node detection information between the network node device and other network node devices, and send the node detection information to the controller;
  • the isolation space dividing unit is configured to divide a plurality of network isolation spaces in the network node device, and the network isolation space is bound to the customer identifier; wherein, the routing policy stored in the network node device is sent to In the network isolation space with the same customer ID.
  • the present application also provides a network node device.
  • the network node device includes a memory and a processor.
  • the memory is used to store a computer program.
  • the computer program is executed by the processor, the information routing method described above can be implemented. .
  • the controller calculates the optimal path for data message transmission based on the path decision algorithm based on the node detection information reported by the network node device, and generates the corresponding routing strategy, and then the controller will route The strategy is distributed to each network node device on the optimal path.
  • the network node device only needs to forward the data message according to the received routing strategy, no longer need to care about its own routing table and other traditional entries, and realize the entire network control.
  • the separation of the plane and the data plane simplifies the deployment of the network, and through the adjustment of the path decision algorithm, the technical solution provided by this application can also flexibly adjust the optimal path of data transmission according to the actual business needs of the user, and improve the work of the network effectiveness.
  • the technical solution provided by this application can also realize the multiplexing of network node equipment through network isolation technology, improve the utilization rate of the system, and enable the node detection information obtained by the network node equipment to be reused by different enterprise users, thereby reducing the network
  • the number of detection request messages sent by the node device in the network reduces the network load.
  • each implementation manner can be implemented by software plus a necessary general hardware platform, and of course, it can also be implemented by hardware.
  • the above technical solutions can be embodied in the form of software products, which can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., include a number of instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiment.

Landscapes

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

Abstract

本发明公开了一种路由分发方法及控制器、信息路由方法及网络节点设备，所述方法应用于控制器中，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述方法包括：响应于第一客户端指向第二客户端的首次访问请求；基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略；向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。本申请提供的技术方案，可以根据用户的实际业务需求，灵活调整数据传输的最优路径，提高网络的工作效率。

Description

路由分发方法及控制器、信息路由方法及网络节点设备 技术领域
本发明涉及互联网技术领域，特别涉及一种路由分发方法及控制器、信息路由方法及网络节点设备。
背景技术
随着企业规模的增长以及互联网业务的发展，大型企业需要解决公司总部、IDC、分支机构等之间的网络互联问题。传统的企业广域网解决方案是通过在Internet中使用VPN或部署专线来解决。然而，Internet网络抖动大、高峰拥塞等问题，会导致用户业务交互不畅，部署专线虽然可以提供良好的链路保障，但面临成本高、分支机构接入不灵活、开通业务周期长等问题。
鉴于此，有必要提供一种新的信息路由方法及网络节点设备、路由分发方法及控制器以解决上述不足。
发明内容
本申请的目的在于提供一种路由分发方法及控制器、信息路由方法及网络节点设备，可以简化网络的部署，提高网络的工作效率。
为实现上述目的，本申请一方面提供一种路由分发方法，所述方法应用于控制器中，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述方法包括：响应于第一客户端指向第二客户端的首次访问请求；基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略；向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
为实现上述目的，本申请另一方面还提供一种控制器，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述控制器包括：访问接收单元，用于接收第一客户端指向第二客户端的首次访问请求；路由规划单元，用于基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径 对应的路由策略；路由分发单元，用于向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
为实现上述目的，本申请一方面提供一种信息路由方法，所述方法应用于网络节点设备中，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识，所述方法包括：接收上游节点发来的数据报文，并识别所述数据报文中携带的目标客户标识，以及查询具备所述目标客户标识的目标路由策略；提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点；将所述数据报文转发至所述下游节点。
为实现上述目的，本申请另一方面还提供一种网络节点设备，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识，所述网络节点设备包括：目标用户识别单元，用于接收上游节点发来的数据报文，并识别所述数据报文中携带的目标用户信息，以及查询具备所述目标客户标识的目标路由策略；节点管理单元，用于提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点；报文转发单元，用于将所述数据报文转发至所述下游节点。
由此可见，本申请提供的技术方案，控制器通过网络节点设备上报的节点探测信息，基于路径决策算法计算出数据报文传输的最优路径，并生成对应的路由策略，然后控制器将路由策略分发至最优路径上的各个网络节点设备，网络节点设备只需要按照接收到的路由策略对数据报文进行转发即可，不用再关心自身的路由表等传统表项，实现了整个网络控制平面和数据平面的分离，简化网络的部署，并且通过对路径决策算法的调整，本申请提供的技术方案还可以根据用户的实际业务需求，灵活调整数据传输的最优路径，提高网络的工作效率。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，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的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施方式中SDN网络架构示意图；
图2是本发明实施方式中路由分发方法的流程图；
图3是本发明实施方式中网络拓扑示意图；
图4是本发明实施方式中控制器的功能模块示意图；
图5是本发明实施方式中控制器的结构示意图；
图6是本发明实施方式中信息路由方法的流程图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
为便于理解本申请中涉及到的控制平面和数据平面分离的内容，下面对其进行简要介绍。传统网络是分布式控制的架构，网络中的每台节点设备(路由器、交换机等)都包含独立的控制平面和数据平面，用于协议计算的控制平面和报文转发的数据平面位于同一台节点设备中，当路由计算和网络拓扑发生变化后，每台节点设备都要重新进行路由计算过程，这就导致传统网络不能灵活调整数据的传输路径，而且各个节点设备之间的网络协议复杂，实现机制各不相同，导致部署及更新网络业务时运维复杂。SDN(Software Defined Network，软件定义网络)是对传统网络架构的一次重构，由原来分布式控制的网络架构重构为集中控制的网络架构，SDN将节点设备上的控制层面分离出来，由集中的控制器管理，无须依赖底层节点设备，屏蔽了来自底层节点设备的差异，而控制权是完全开放的，用户可以自定义任何想实现的网络路由和传输规则策略。
请参阅图1，为SDN网络架构示意图。
在SDN中网络架构分为三层：协同应用层、控制层和转发层。协同应用层主要用于管理及运行用户开发的各种应用程序；控制层是整个系统的控制中心，控制层通过南向控制接口和转发层交互，通过北向业务接口和协同应用层交互，其实现的主体是控制器，控制器负责网络的内部交换路径和边界业务路由的生成，并负责处理网络状态变化事件，当网络发生状态变化，比如链路故障、节点故障、网络拥塞时，控制器会根据这些网络状态的变化信息调整网络交换路径和业务路由，使网络始终处于一个正常的工作状态；转发层基于传统网络中的各种网络节点设备组成基础转发网络，转发层一方面向控制层上报网络资源信息和状态，另一方面接收控制层下发的路由策略，其实现的主体是SDN转发器(即支持SDN的各种网络节点设备)，SDN转发器负责执行数据报文的转发， 转发过程所需要的转发表项由控制层生成，SDN转发器是系统的执行单元，本身不做决策。
由此可见，SDN可以通过控制器中的软件平台实现网络节点设备的可编程化控制，实现对网络资源的按需调配，SDN转发器只负责单纯的数据报文转发。控制器具有链路发现、拓扑管理的功能，其中链路发现和拓扑管理主要是控制器利用南向控制接口的上行通道对底层网络节点设备上报的信息进行统一监控和统计。
需要特别指出的是，下文中提及的控制器，其已经创建并储存有完整的网络拓扑图，并且下文中提及的各个客户端以及各个网络节点设备都存在于上述网络拓扑图中，各个客户端、各个网络节点设备以及控制器之间是可以连通的。
请参阅图2，本申请提供的路由分发方法应用于控制器中，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息。
在本实施方式中，网络中的任一网络节点设备可以向与其连接的其它网络节点设备发送探测请求报文，并根据上述其它网络节点设备返回的探测应答报文，获取该网络节点设备与上述其它网络节点设备之间的节点探测信息。
该网络节点设备在获取到上述节点探测信息后，可以利用南向控制接口的上行通道将上述节点探测信息发送至控制器，以使得控制器储存上述节点探测信息。需要特别指出的是，网络节点设备可以根据预设时间值，主动定时发送探测请求报文，也可以在网络拓扑结构发生改变时，根据控制器下发的探测信息更新指令，被动发送探测请求报文，本申请对此不作限制。
在本实施方式中，网络中的任一客户端还可以向与其连接的网络节点设备发送探测请求报文，并根据上述网络节点设备返回的探测应答报文，获取该客户端与上述网络节点设备之间的用户探测信息。
该客户端在获取到上述用户探测信息后，可以利用南向控制接口的上行通道将上述用户探测信息发送至控制器，以使得控制器储存上述用户探测信息。需要特别指出的是，客户端可以根据预设时间值，主动定时发送探测请求报文，也可以在网络拓扑结构发生改变时，根据控制器下发的探测信息更新指令，被动发送探测请求报文，本申请对此不作限制。
在实际应用中，网络中的某一个客户端可能同时与多个网络节点设备相连接，上述多个网络节点设备可能与该客户端处于同一个局域网，也可能与该客 户端处于不同的局域网，在这种情况下，客户端可以只向与其处于同一局域网中的网络节点设备发送探测请求报文，以减少客户端在网络中发送的探测请求报文的数量，进而减少网络负载。需要特别指出的是，客户端也可以只向与其地理位置相近且处于同一运营商网络下的网络节点设备发送探测请求报文，本申请对此不作限制。
需要特别指出的是，上述节点探测信息以及用户探测信息中至少包括网络时延、网络抖动和网络丢包率等网络质量信息，在实际应用中，随着技术手段的不断丰富和应用场景的变换，节点探测信息和用户探测信息中包含的内容可以更多或者更少，本申请对此并不做限定。
需要特别指出的是，网络中的所有网络节点设备都将进行上述发送探测请求报文和发送节点探测信息的动作，网络中的所有客户端都将进行上述发送探测请求报文和发送用户探测信息的动作，以使得控制器可以储存所有网络节点设备发送的节点探测信息和所有客户端发送的用户探测信息，控制器根据储存的网络拓扑图，进而可以获取网络拓扑图中所有节点设备之间的网络质量信息。
需要特别指出的是，当本申请中的路由分发方法应用于企业广域网时，各个客户端还具备有客户标识，上述客户标识与企业广域网标识相绑定，上述客户标识可以用于表明某一客户端是否与其它客户端处于同一企业广域网中。相应的，当某一客户端向控制器发送用户探测信息时，该客户端还需要将其客户标识发送至控制器。
本申请提供的路由分发方法可以包括以下步骤。
S101：接收第一客户端发来的指向第二客户端的访问请求。
在本实施方式中，第一客户端需要访问第二客户端，此时第一客户端发送的目的地址为第二客户端的访问请求首先会到达控制器，显然，上述访问请求中携带有第一客户端的地址以及第二客户端的地址，以使得控制器根据上述第一客户端的地址以及第二客户端的地址，确定访问请求的源地址和目的地址。
在一个实施方式中，当上述第一客户端和上述第二客户端位于同一企业广域网时，上述第一客户端和第二客户端还将具备有相同的客户标识，上述访问请求中还将携带有上述客户标识，具备相同客户标识的第一客户端和第二客户端可以位于同一企业广域网内。
在实际应用中，当控制器接收到第一客户端指向第二客户端的访问请求时，可以先确定当前是否已经存在对应的路由策略，如果没有，则表明控制器是首 次接收到该访问请求，此时，可以按照下列对所述首次请求进行响应；如果有，则表明控制器不是首次接收到该访问请求，此时，可以按照已有的路由策略进行转发。同时，可以记录访问请求的请求次数，以统计确认出热点访问请求，并可以针对热点的访问请求，对相应的路由策略进行实时更新。
S102：响应于第一客户端指向第二客户端的首次访问请求，基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略。
在本实施方式中，控制器可以根据储存的节点探测信息和用户探测信息，利用最短路径算法计算出上述第一客户端和上述第二客户端之间的最优路径，上述最优路径中包含有数据报文从上述第一客户端发送至上述第二客户端时所要经过的所有网络节点设备以及各个网络节点设备之间的上下游连接关系。控制器根据上述最优路径生成对应的路由策略，上述路由策略中至少包括用户信息和节点信息。上述节点信息至少包括数据报文从上述第一客户端发送至上述第二客户端时所要经过的所有网络节点设备的地址信息。上述用户信息中至少包括第一客户端和第二客户端的客户标识，以及用于表征上述最优路径起始端及目的端的地址信息。
需要特别指出的是，当第一客户端发出指向第二客户端的数据报文时，上述数据报文可以按照上述路由策略最终抵达第二客户端。
S103：向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
控制器生成最优路径对应的路由策略后，可以利用南向控制接口的下行通道，将上述路由策略发送至上述最优路径中的所有网络节点设备和/或客户端。
在一个实施方式中，当控制器生成上述最优路径对应的路由策略后，控制器可以将完整的路由策略统一分发至上述最优路径中的所有网络节点设备和/或客户端，以减少控制器的计算负载。针对最优路径中的某一个网络节点设备或者客户端而言，其接收到的路由策略中既包含有该网络节点设备或者该客户端自身需要执行的路由指令，也包含有上述最优路径中其它网络节点设备和客户端需要执行的路由指令。
在另一个实施方式中，当控制器生成上述最优路径对应的路由策略后，控制器还可以根据上述最优路径中的网络节点设备的标识和客户端的标识，将上述路由策略划分为路由子策略，并根据上述网络节点设备的标识和上述客户端 的标识，将上述路由子策略分发至对应的网络节点设备或者客户端，上述路由子策略中只包含有上述网络节点设备或者上述客户端需要执行的路由指令，以减少控制器下发路由策略时的网络负载，并提高网络节点设备或者客户端转发数据报文时的效率。需要特别指出的是，上述网络节点设备的标识和客户端的标识可以使用设备的MAC地址，也可以由控制器根据网络拓扑图赋予设备唯一的编号来确定，本申请对此不作限制。
在一个实施方式中，由于控制器中储存有完整的网络拓扑图，因此控制器可以根据网络拓扑图获取上述第一客户端和上述第二客户端之间的所有连接路径(即候选路径集合)，上述连接路径中包含有数据报文从上述第一客户端发送至上述第二客户端时所要经过的所有网络节点设备。
针对某一条候选路径而言，控制器可以根据该候选路径中包含的所有网络节点设备以及各个网络节点设备之间的上下游连接关系，在储存的节点探测信息中识别出彼此直接连接的两个网络节点设备之间的节点探测信息，并在储存的用户探测信息中识别出第一客户端和与第一客户端直接连接的网络节点设备之间的用户探测信息，以及第二客户端和与第二客户端直接连接的网络节点设备之间的用户探测信息，根据上述节点探测信息和上述用户探测信息，控制器可以计算出该候选路径对应的路径指标值。对候选路径集合中的所有候选路径执行相同的操作，控制器便可以计算出候选路径集合中的所有候选路径对应的路径指标值，并基于上述路径指标值，确定最优路径。
控制器在计算某一条候选路径对应的路径指标值时，可以基于节点探测信息和用户探测信息，获取该候选路径对应的网络质量参数，并根据上述网络质量参数和与客户标识相绑定的权值参数，计算该候选路径的路径指标值，上述权值参数可以预先设置在控制器中。在实际应用中，由于不同的用户可能具有不同的业务需求，因此针对不同的用户可以设置不同的权值参数，并将上述权值参数与客户标识相绑定，用于区分不同用户的权值参数。
需要特别指出的是，上述网络质量参数至少包括时延参数和/或抖动参数和/或丢包率，上述权值参数至少包括时延权值和/或抖动权值和/或丢包率权值，上述权值参数可以根据用户的业务类型进行设定。
在实际应用中，一个企业广域网下可能同时拥有多个分支局域网，各个分支局域网中的终端设备不具备公网IP地址，但具备不同的内网网段，以使得一个分支局域网中的终端设备可以访问另一个分支局域网中的终端设备，此时上 述客户端可以是客户前置设备(Customer premises equipment，CPE)，用于连接局域网和外网。需要特别指出的是，此时可以在上述客户端和与其直接连接的网络节点设备之间建立专用网络，上述专用网络可以基于公网链路，利用VPN(Virtual Private Network，虚拟专用网络)实现客户端与网络节点设备之间的信息交互；上述专用网络还可以使用二层专线技术，在客户端和网络节点设备之间建立专线连接，进而实现客户端与网络节点设备之间的信息交互，本申请对此不作限制。
在一个实施方式中，当上述客户端接收到某一局域网中的起始终端设备发来的，指向另一局域网中的目标终端设备的数据报文时，客户端可以将上述起始终端设备的内网网段，以及上述目标终端设备的内网网段发送至控制器，以使得控制器可以根据上述起始终端设备的内网网段，以及上述目标终端设备的内网网段规划路由路径，保证上述数据报文可以正确转发至上述目标终端设备。
下面举例说明本申请提供的路由分发方法的一种实现方式。
请参阅图3，假设客户端1(记为CPE1)与客户端2(记为CPE2)位于同一企业广域网中，客户端3(记为CPE3)与客户端4(记为CPE4)位于另一企业广域网中，CPE1与CPE2具备相同的客户标识500100，CPE3与CPE4具备相同的客户标识500125，网络节点设备1(记为POP1)、网络节点设备2(记为POP2)与CPE1处于同一局域网，网络节点设备N(记为POPN)与CPE2处于同一局域网，CPE1、CPE2、CPE3、CPE4、POP1、POP2和POPN都位于控制器的网络拓扑图中。控制器根据网络拓扑图以及客户标识500100可以确定CPE1和CPE2之间的候选路径为CPE1-POP1-POPN-CPE2和CPE1-POP2-POPN-CPE2。
由于控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，因此控制器中必然储存有POP1和POP2之间的节点探测信息(包括POP1和POP2之间的时延参数D1、抖动参数S1和丢包率L1)、POP1和POPN之间的节点探测信息(包括POP1和POPN之间的时延参数D2、抖动参数S2和丢包率L2)、POP2和POPN之间的节点探测信息(包括POP2和POPN之间的时延参数D3、抖动参数S3和丢包率L3)、CPE1和POP1之间的用户探测信息(包括CPE1和POP1之间的时延参数D4、抖动参数S4和丢包率L4)、CPE1和POP2之间的用户探测信息(包括CPE1和POP2之间的时延参数D5、抖动参数S5和丢包率L5)，以及CPE2和POPN之间的用户探测信息(包括CPE2和POPN之间的时延参数D6、抖动参数S6和丢包率L6)。
控制器根据预设的与客户标识500100相绑定的权值参数(假设时延权值为α、抖动权值为β、丢包率权值为γ)，利用路径决策算法，计算每条候选路径的路径指标值，其中：
CPE1-POP1-POPN-CPE2路径的路径指标值：
In1＝α*(D4+D2+D6)+β*(S4+S2+S6)+γ*(L4+L2+L6)*100，
CPE1-POP2-POPN-CPE2路径的路径指标值：
In2＝α*(D5+D3+D6)+β*(S5+S3+S6)+γ*(L5+L3+L6)*100，
若In1小于In2，则表示CPE1-POP1-POPN-CPE2路径优于CPE1-POP2-POPN-CPE2路径，控制器将选择CPE1-POP1-POPN-CPE2路径为最优路径，并生成CPE1-POP1-POPN-CPE2路径对应的路由策略，并将生成的上述路由策略分发至CPE1、POP1、POPN和CPE2，上述路由策略由CPE1、POP1、POPN和CPE2各自所要执行的路由指令组成。
需要特别指出的是，控制器可以将完整的路由策略统一分发至CPE1、POP1、POPN和CPE2，也可以根据CPE1、POP1、POPN和CPE2各自的MAC地址，将上述路由策略划分为路由子策略，并根据MAC地址将上述路由子策略分别发送至CPE1、POP1、POPN和CPE2，上述路由子策略中只包含有上述网络节点设备或者上述客户端需要执行的路由指令。
现以CPE1-POP1-POPN-CPE2路径中的POP1举例说明控制器如何划分以及分发路由子策略。由于网络中的数据报文是可以双向流动的，因此POP1接收到的路由子策略包含有两个方向的路由指令，控制器可以在上述路由策略中查询到POP1所要执行的路由指令，上述POP1所要执行的路由指令包括(目的地址CPE1，下一跳节点CPE1)；(目的地址CPE2，下一跳节点POPN),然后控制器只需将上述路由指令发送至POP1，以使得POP1接收到的路由指令只包含其自身需要执行的路由指令。需要特别指出的是，上述时延权值α、抖动权值β以及丢包率权值γ，可以根据用户的实际业务需要进行设定。例如，客户标识为500100的用户，其业务为时延敏感型业务，在设定权值参数时，相应的可以增加时延权值α的值；客户标识为500125的用户，其业务为抖动敏感型业务，在设定权值参数时，相应的可以增加抖动权值β的值。
需要特别指出的是，上述列举的路径决策算法只是一种示例，并不是对路径决策算法的限定，基于本申请的思想，本领域的技术人员也可以采用其它路径决策算法。
在实际应用中，同一个网络节点设备中可以通过网络隔离技术，划分为多个网络隔离空间，各个网络隔离空间之间可以形成资源隔离，并且各个网络隔离空间可以被不同的企业用户租用，上述网络隔离空间可以与客户标识相绑定。这样，同一个网络节点设备，可以同时向多个企业用户提供数据转发服务，并且通过客户标识来区分位于同一个网络节点设备中的不同企业用户，可以保证不同企业用户的路由策略互不影响，提高系统的利用率，相应的，网络节点设备获取的节点探测信息也可以被不同的企业用户复用，进而减少网络节点设备在网络中发送的探测请求报文的数量，减轻网络负载。
请参阅图4，本申请还提供一种控制器，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述控制器包括：
访问接收单元，用于接收第一客户端发来的指向第二客户端的访问请求，并判断所说访问请求是否为首次访问请求，并将首次访问请求发送至路由规划单元；
路由规划单元，用于基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略，以对接收到的首次访问请求进行响应；
路由分发单元，用于向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
在一个实施方式中，上述路由规划单元还包括：
子路由规划模块，用于根据所述最优路径中的网络节点设备的标识和客户端的标识将所述路由策略划分为路由子策略，并将所述路由子策略分发至对应的所述网络节点设备或者客户端。
请参阅图5，本申请还提供一种控制器，所述控制器包括存储器和处理器，所述存储器用于存储计算机程序，当所述计算机程序被所述处理器执行时，可以实现如上述的路由分发方法。具体地，在硬件层面，该题目分发装置可以包括处理器、内部总线和存储器。所述存储器可以包括内存以及非易失性存储器。处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行。本领域普通技术人员可以理解，图5所示的结构仅为示意，其并不对上述识别装置的结构造成限定。例如，所述题目分发装置还可包括比图5中所示更多或者更少的组件，例如还可以包括其他的处理硬件，如GPU(Graphics Processing Unit，图像处理器)，或者对外通信端口等。当然，除了软件实现方式之外，本申请并 不排除其他实现方式，比如逻辑器件抑或软硬件结合的方式等等。
本实施方式中，所述的处理器可以包括中央处理器(CPU)或图形处理器(GPU)，当然也可以包括其他的具有逻辑处理能力的单片机、逻辑门电路、集成电路等，或其适当组合。本实施方式所述的存储器可以是用于保存信息的记忆设备。在数字系统中，能保存二进制数据的设备可以是存储器；在集成电路中，一个没有实物形式的具有存储功能的电路也可以为存储器，如RAM、FIFO等；在系统中，具有实物形式的存储设备也可以叫存储器等。实现的时候，该存储器也可以采用云存储器的方式实现，具体实现方式，本说明书不做限定。
需要说明的是，本说明书中的控制器，具体的实现方式可以参照方法实施方式的描述，在此不作一一赘述。
请参阅图6，本申请还提供一种信息路由方法，所述方法应用于网络节点设备中，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识。
在本实施方式中，网络节点设备可以同时为多个企业广域网服务，因此上述网络节点设备中储存有控制器下发的多个路由策略，上述多个路由策略具备不同的客户标识。上述客户标识与企业广域网以及企业广域网中的客户端相绑定，可以用于确定具备不同客户标识的客户端所要执行的路由策略，上述客户标识还可以用于表明数据报文的源客户端与目的客户端是否处于同一企业广域网中。
需要特别指出的是，上述路由策略由控制器根据数据报文的源客户端和目的客户端之间的最优路径生成，上述路由策略可以是完整的路由策略，也可以是控制器根据网络节点设备的标识生成的路由子策略。
具体的，所述方法可以包括以下步骤：
S201：接收上游节点发来的数据报文，并识别所述数据报文中携带的目标客户标识，以及查询具备所述目标客户标识的目标路由策略。
在本实施方式中，网络节点设备接收到上游节点转发来的数据报文，上述数据报文中携带有目标客户标识，具体的，上述目标客户标识可以为数据报文的源客户端具备的客户标识。网络节点设备识别出上述目标客户标识，并根据上述目标客户标识，在储存的多个路由策略中查询出具备上述目标客户标识的目标路由策略，上述目标路由策略中包含有下一跳地址对应的下游节点，可以用于向下游节点转发携带有上述目标客户标识的数据报文。
S202:提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点.
在本实施方式中，上述数据报文中还携带有目的地址，具体的，上述目的地址可以为数据报文的目的客户端的地址，网络节点设备可以从上述数据报文中提取出上述目的地址，并根据上述目的地址的指向确定下一跳地址，以及与下一跳地址对应的下游节点。
现以上文中的CPE1-POP1-POPN-CPE2路径，举例说明网络节点设备如何根据目的地址的指向确定下一跳地址，以及与下一跳地址对应的下游节点。由于网络中的数据报文是可以双向流动的，因此网络节点设备在转发数据报文时，首先需要判断接收到的数据报文的流向，上述数据报文的流向可以根据数据报文中携带的目的地址确定，例如，当POP1接收到CPE1发来的数据报文，上述数据报文中携带的目的地址为CPE2的地址，因此POP1可以确定上述数据报文的流向为指向CPE2的方向，进而POP1根据储存的路由策略确定下一跳地址为POPN的地址，对应的下游节点为POPN。S203：将所述数据报文转发至所述下游节点。
在本实施方式中，网络节点设备将根据下一跳地址，将上述数据报文转发至下一跳地址对应的下游节点。
在一个实施方式中，还可以利用网络隔离技术，在网络节点设备中划分出多个网络隔离空间，上述网络隔离空间可以与客户标识相绑定。例如，可以使用namespace技术在网络节点设备中划分出多个网络隔离空间，本申请对实现网络隔离的技术不作限制。当上述网络隔离空间与客户标识相绑定后，网络节点设备便可以根据路由策略具备的不同客户标识，将路由策略发送至具备相同客户标识的网络隔离空间中。通过在一个网络节点设备中划分出多个网络隔离空间，可以使该网络节点设备被具备不同客户标识的客户端(即不同的企业用户)复用，提高系统的利用率，相应的，网络节点设备获取的节点探测信息也可以被不同的企业用户复用，进而减少网络节点设备在网络中发送的探测请求报文的数量，减轻网络负载。
在一个实施方式中，上述网络节点设备还可以向与其连接的其它网络节点设备发送探测请求报文，并根据上述其它网络节点设备返回的探测应答报文，获取该网络节点设备与上述其它网络节点设备之间的节点探测信息。该网络节点设备在获取到上述节点探测信息后，可以利用南向控制接口的上行通道将上 述节点探测信息发送至控制器，以使得控制器储存上述节点探测信息。需要特别指出的是，网络节点设备可以根据预设时间值，主动定时发送探测请求报文，也可以在网络拓扑结构发生改变时，根据控制器下发的探测信息更新指令，被动发送探测请求报文，本申请对此不作限制。
需要特别指出的是，当与网络节点设备直接连接的客户端为客户前置设备时，还可以在上述网络节点设备和上述客户端之间建立专用网络。上述专用网络可以基于公网链路，利用VPN(Virtual Private Network，虚拟专用网络)实现客户端与网络节点设备之间的信息交互；上述专用网络还可以使用二层专线技术，在客户端和网络节点设备之间建立专线连接，进而实现客户端与网络节点设备之间的信息交互，本申请对此不作限制。
本申请还提供一种网络节点设备，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识，所述网络节点设备包括：
目标用户识别单元，用于接收上游节点发来的数据报文，并识别所述数据报文中携带的目标用户信息，以及查询具备所述目标客户标识的目标路由策略；
节点管理单元，用于提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点；
报文转发单元，用于将所述数据报文转发至所述下游节点。
在一个实施方式中，上述网络节点设备还包括：
探测信息上报单元，用于获取所述网络节点设备与其它网络节点设备之间的节点探测信息，并将所述节点探测信息发送至所述控制器；
隔离空间划分单元，用于在所述网络节点设备中划分多个网络隔离空间，所述网络隔离空间与所述客户标识相绑定；其中，所述网络节点设备中存储的路由策略被发送至具备相同所述客户标识的网络隔离空间中。
本申请还提供一种网络节点设备，所述网络节点设备包括存储器和处理器，所述存储器用于存储计算机程序，所述计算机程序被所述处理器执行时，可以实现如上述的信息路由方法。
需要说明的是，本说明书中的网络节点设备，具体的实现方式可以参照方法实施方式的描述，在此不作一一赘述。
由此可见，本申请提供的技术方案，控制器通过网络节点设备上报的节点探测信息，基于路径决策算法计算出数据报文传输的最优路径，并生成对应的路由策略，然后控制器将路由策略分发至最优路径上的各个网络节点设备，网 络节点设备只需要按照接收到的路由策略对数据报文进行转发即可，不用再关心自身的路由表等传统表项，实现了整个网络控制平面和数据平面的分离，简化了网络的部署，并且通过对路径决策算法的调整，本申请提供的技术方案还可以根据用户的实际业务需求，灵活调整数据传输的最优路径，提高网络的工作效率。本申请提供的技术方案还可以通过网络隔离技术，实现网络节点设备的复用，提高了系统的利用率，并且使网络节点设备获取的节点探测信息可以被不同的企业用户复用，进而减少网络节点设备在网络中发送的探测请求报文的数量，减轻网络负载。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件来实现。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (14)

  1. 一种路由分发方法，其特征在于，所述方法应用于控制器中，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述方法包括：
    响应于第一客户端指向第二客户端的首次访问请求；
    基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略；
    向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
  2. 根据权利要求1所述的方法，其特征在于，在生成所述最优路径对应的路由策略后，所述方法还包括：
    根据所述最优路径中的网络节点设备的标识和客户端的标识将所述路由策略划分为路由子策略，并将所述路由子策略分发至对应的所述网络节点设备或者客户端。
  3. 根据权利要求1所述的方法，其特征在于，在所述第一客户端和所述第二客户端之间规划最优路径包括：
    获取所述第一客户端和所述第二客户端之间的候选路径集合；
    根据所述节点探测信息和所述用户探测信息，计算所述候选路径集合中的候选路径的路径指标值；
    基于所述路径指标值，确定所述最优路径。
  4. 根据权利要求3所述的方法，其特征在于，计算所述候选路径集合中的候选路径的路径指标值包括：
    基于所述节点探测信息和所述用户探测信息，获取所述候选路径对应的网络质量参数；
    根据所述网络质量参数和与所述客户标识相绑定的权值参数，计算所述候选路径的路径指标值。
  5. 根据权利要求4所述的方法，其特征在于，所述网络质量参数至少包括 时延参数和/或抖动参数和/或丢包率，所述权值参数至少包括时延权值和/或抖动权值和/或丢包率权值。
  6. 一种控制器，其特征在于，所述控制器储存有各个网络节点设备发送的节点探测信息以及各个客户端发送的用户探测信息，所述控制器包括：
    访问接收单元，用于接收第一客户端指向第二客户端的首次访问请求；
    路由规划单元，用于基于所述节点探测信息和所述用户探测信息，在所述第一客户端和所述第二客户端之间规划最优路径，并生成所述最优路径对应的路由策略；
    路由分发单元，用于向位于所述最优路径中的网络节点设备和/或客户端分发所述路由策略。
  7. 根据权利要求6所述的控制器，其特征在于，所述路由规划单元还包括：
    子路由规划模块，用于根据所述最优路径中的网络节点设备的标识和客户端的标识将所述路由策略划分为路由子策略，并将所述路由子策略分发至对应的所述网络节点设备或者客户端。
  8. 一种控制器，其特征在于，所述控制器包括存储器和处理器，所述存储器用于存储计算机程序，当所述计算机程序被所述处理器执行时，实现如权利要求1至5中任一权利要求所述的方法。
  9. 一种信息路由方法，其特征在于，所述方法应用于网络节点设备中，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识，所述方法包括：
    接收上游节点发来的数据报文，并识别所述数据报文中携带的目标客户标识，以及查询具备所述目标客户标识的目标路由策略；
    提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点；
    将所述数据报文转发至所述下游节点。
  10. 根据权利要求8所述的方法，其特征在于，在所述网络节点设备中划分 有多个网络隔离空间，所述网络隔离空间与所述客户标识相绑定；其中，所述网络节点设备中存储的路由策略被发送至具备相同所述客户标识的网络隔离空间中。
  11. 根据权利要求8所述的方法，其特征在于，所述方法还包括：
    获取所述网络节点设备与其它网络节点设备之间的节点探测信息；
    将所述节点探测信息发送至所述控制器。
  12. 一种网络节点设备，其特征在于，所述网络节点设备中储存有控制器下发的多个路由策略，所述路由策略中具备客户标识，所述网络节点设备包括：
    目标用户识别单元，用于接收上游节点发来的数据报文，并识别所述数据报文中携带的目标用户信息，以及查询具备所述目标客户标识的目标路由策略；
    节点管理单元，用于提取所述数据报文的目的地址，并从所述目标路由策略中确定所述目的地址指向的下游节点；
    报文转发单元，用于将所述数据报文转发至所述下游节点。
  13. 根据权利要求12所述的网络节点设备，其特征在于，所述网络节点设备还包括：
    探测信息上报单元，用于获取所述网络节点设备与其它网络节点设备之间的节点探测信息，并将所述节点探测信息发送至所述控制器；
    隔离空间划分单元，用于在所述网络节点设备中划分多个网络隔离空间，所述网络隔离空间与所述客户标识相绑定；其中，所述网络节点设备中存储的路由策略被发送至具备相同所述客户标识的网络隔离空间中。
  14. 一种网络节点设备，其特征在于，所述网络节点设备包括存储器和处理器，所述存储器用于存储计算机程序，当所述计算机程序被所述处理器执行时，实现如权利要求9至11中任一权利要求所述的方法。
PCT/CN2019/110770 2019-07-18 2019-10-12 路由分发方法及控制器、信息路由方法及网络节点设备 WO2021007963A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP19937704.5A EP3993328A1 (en) 2019-07-18 2019-10-12 Route distribution method and controller, information routing method and network node device
US17/627,696 US20220329514A1 (en) 2019-07-18 2019-10-12 Routing distribution method and controller, information routing method, and network node device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910650501.7 2019-07-18
CN201910650501.7A CN112242949A (zh) 2019-07-18 2019-07-18 路由分发方法及控制器、信息路由方法及网络节点设备

Publications (1)

Publication Number Publication Date
WO2021007963A1 true WO2021007963A1 (zh) 2021-01-21

Family

ID=74168265

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110770 WO2021007963A1 (zh) 2019-07-18 2019-10-12 路由分发方法及控制器、信息路由方法及网络节点设备

Country Status (4)

Country Link
US (1) US20220329514A1 (zh)
EP (1) EP3993328A1 (zh)
CN (1) CN112242949A (zh)
WO (1) WO2021007963A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114726776A (zh) * 2022-01-30 2022-07-08 阿里巴巴（中国）有限公司 内容分发网络cdn调度方法、装置、设备及介质
CN115379016A (zh) * 2022-08-22 2022-11-22 深信服科技股份有限公司 资源访问方法、访问服务平台、装置、设备及存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115150312B (zh) * 2021-03-31 2024-05-17 华为技术有限公司 一种路由方法及设备
CN113285941B (zh) * 2021-05-18 2023-09-08 中国联合网络通信集团有限公司 企业外网系统及网络请求处理方法
CN113300914A (zh) * 2021-06-28 2021-08-24 北京字跳网络技术有限公司 网络质量监测方法、装置、系统、电子设备和存储介质
CN113542064B (zh) * 2021-07-13 2023-02-28 北京字跳网络技术有限公司 网络路径确定方法、装置、电子设备、介质及程序产品
CN113872824B (zh) * 2021-08-19 2023-04-07 网宿科技股份有限公司 Cdn网络质量检测方法、系统、服务器及存储介质
CN115297048B (zh) * 2022-07-07 2023-04-07 北京瑞祺皓迪技术股份有限公司 一种基于光纤网络的路由路径生成方法及装置
CN115277541B (zh) * 2022-08-01 2024-01-26 明阳产业技术研究院(沈阳)有限公司 基于IPv6的通信方法及相关设备
CN118018472A (zh) * 2022-11-08 2024-05-10 中兴通讯股份有限公司 一种数据传输处理方法、装置、存储介质及电子装置
CN115883444A (zh) * 2022-12-23 2023-03-31 北京云端智度科技有限公司 基于最优路径的网络数据传输方法、系统、介质及设备
CN117675681A (zh) * 2023-12-08 2024-03-08 广州易云信息技术有限公司 一种多渠道路由策略管理方法、装置及电子设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102857416A (zh) * 2012-09-18 2013-01-02 中兴通讯股份有限公司 一种实现虚拟网络的方法和虚拟网络
US20130223226A1 (en) * 2012-02-29 2013-08-29 Dell Products, Lp System and Method for Providing a Split Data Plane in a Flow-Based Switching Device
CN104660700A (zh) * 2015-03-03 2015-05-27 网宿科技股份有限公司 一种内容分发网络的方法和系统
CN105490945A (zh) * 2014-09-15 2016-04-13 上海贝尔股份有限公司 一种用于在控制面中控制转发数据的方法和装置
CN109787868A (zh) * 2019-03-18 2019-05-21 网宿科技股份有限公司 一种选取路由路径的方法、系统及服务器

Family Cites Families (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3546764B2 (ja) * 1999-07-02 2004-07-28 日本電気株式会社 ネットワークに備えられた負荷分散サーバ及び負荷分散サーバを備えるノード
DE60233759D1 (de) * 2002-05-08 2009-10-29 Nokia Corp Verfahren zum verteilen von netzwerkparametern unter netzwerkknoten
US8493869B2 (en) * 2009-10-19 2013-07-23 Cisco Technology, Inc. Distributed constraints-based inter-domain network traffic management
JP5375661B2 (ja) * 2010-02-24 2013-12-25 富士通株式会社 経路割当装置および経路割当方法
JP5919046B2 (ja) * 2012-03-13 2016-05-18 株式会社日立製作所 パス計算方法
CN103873368B (zh) * 2012-12-11 2017-09-15 新华三技术有限公司 以太网报文转发方法和接入设备
JP6007799B2 (ja) * 2013-01-16 2016-10-12 富士通株式会社 集中管理型網制御システム
CN104579961B (zh) * 2013-10-11 2018-09-07 中国移动通信集团公司 数据报文的调度方法及装置
US9537753B2 (en) * 2014-03-03 2017-01-03 Cisco Technology, Inc. Opaque profile identifiers for path computation element protocol
US10484276B2 (en) * 2014-08-01 2019-11-19 Hitachi, Ltd. Route resolution system and route resolution method
US10148578B2 (en) * 2014-10-17 2018-12-04 Ciena Corporation Optical and packet path computation and selection systems and methods
US9800507B2 (en) * 2015-02-10 2017-10-24 Verizon Patent And Licensing Inc. Application-based path computation
US20160254984A1 (en) * 2015-02-27 2016-09-01 Argela Yazilim ve Bilisim Teknolojileri San. ve Tic. A.S. Method and system for delivering service-enabled flow paths across multiple domains in sdn networks
CN104780095A (zh) * 2015-04-30 2015-07-15 杭州华三通信技术有限公司 一种sdn网络中的路径探测方法和装置
EP3229405B1 (en) * 2015-12-31 2020-07-15 Huawei Technologies Co., Ltd. Software defined data center and scheduling and traffic-monitoring method for service cluster therein
US10581723B2 (en) * 2016-04-01 2020-03-03 Futurewei Technologies, Inc. PCEP extension for PCECC support of distributed computing, multiple services, and inter-domain routing
CN106100990B (zh) * 2016-06-06 2019-04-05 中国电子科技集团公司第三十研究所 一种基于sdn架构的通信网络多维度资源调控方法和系统
CN106130767B (zh) * 2016-09-23 2020-04-07 深圳灵动智网科技有限公司 一种业务路径故障监控及故障解决的系统和方法
US10298488B1 (en) * 2016-09-30 2019-05-21 Juniper Networks, Inc. Path selection and programming of multiple label switched paths on selected paths of multiple computed paths
CN106789648B (zh) * 2016-12-15 2019-09-03 南京邮电大学 基于内容存储与网络状况的软件定义网络路由决策方法
CN108306783A (zh) * 2017-12-26 2018-07-20 深圳市泰信通信息技术有限公司 基于sdn的转发方法、装置及计算机可读存储介质
CN110011915A (zh) * 2018-01-04 2019-07-12 杭州达乎科技有限公司 Sdn网络架构、基于sdn的流量转发控制方法及装置
WO2019210769A1 (en) * 2018-05-03 2019-11-07 Huawei Technologies Co., Ltd. Explicit routing with network function encoding
US10999189B2 (en) * 2018-11-20 2021-05-04 Cisco Technology, Inc. Route optimization using real time traffic feedback

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130223226A1 (en) * 2012-02-29 2013-08-29 Dell Products, Lp System and Method for Providing a Split Data Plane in a Flow-Based Switching Device
CN102857416A (zh) * 2012-09-18 2013-01-02 中兴通讯股份有限公司 一种实现虚拟网络的方法和虚拟网络
CN105490945A (zh) * 2014-09-15 2016-04-13 上海贝尔股份有限公司 一种用于在控制面中控制转发数据的方法和装置
CN104660700A (zh) * 2015-03-03 2015-05-27 网宿科技股份有限公司 一种内容分发网络的方法和系统
CN109787868A (zh) * 2019-03-18 2019-05-21 网宿科技股份有限公司 一种选取路由路径的方法、系统及服务器

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114726776A (zh) * 2022-01-30 2022-07-08 阿里巴巴（中国）有限公司 内容分发网络cdn调度方法、装置、设备及介质
CN114726776B (zh) * 2022-01-30 2024-01-30 阿里巴巴（中国）有限公司 内容分发网络cdn调度方法、装置、设备及介质
CN115379016A (zh) * 2022-08-22 2022-11-22 深信服科技股份有限公司 资源访问方法、访问服务平台、装置、设备及存储介质

Also Published As

Publication number Publication date
EP3993328A1 (en) 2022-05-04
US20220329514A1 (en) 2022-10-13
CN112242949A (zh) 2021-01-19

Similar Documents

Publication Publication Date Title
WO2021007963A1 (zh) 路由分发方法及控制器、信息路由方法及网络节点设备
US10742556B2 (en) Tactical traffic engineering based on segment routing policies
JP7417825B2 (ja) スライスベースルーティング
EP3399703B1 (en) Method for implementing load balancing, apparatus, and network system
EP2911348B1 (en) Control device discovery in networks having separate control and forwarding devices
US7590074B1 (en) Method and apparatus for obtaining routing information on demand in a virtual private network
US20160134591A1 (en) VPN Implementation Processing Method and Device for Edge Device
US10630508B2 (en) Dynamic customer VLAN identifiers in a telecommunications network
AlSaeed et al. Multicasting in software defined networks: A comprehensive survey
WO2007134551A1 (fr) Procédé et dispositif noeud de réservation de ressources de réseau
WO2018233580A1 (zh) 一种网络中建立转发路径的方法、控制器及系统
WO2015039617A1 (zh) 一种报文处理方法、系统及设备
US11296997B2 (en) SDN-based VPN traffic scheduling method and SDN-based VPN traffic scheduling system
WO2019184653A1 (zh) 链路配置方法和控制器
US10069763B2 (en) Method to establish a non-disruptive communications path between multiple devices
CN114006857A (zh) 路径规划方法及装置
CN109286563B (zh) 一种数据传输的控制方法和装置
CN113300957A (zh) 一种基于段路由的智能骨干网管理调度系统和方法
CN111092824B (zh) 流量管理系统、方法、电子终端、及存储介质
López-Rodríguez et al. A robust SDN network architecture for service providers
WO2021219049A1 (zh) 一种信息上报方法、信息处理方法、装置及设备
WO2016086721A1 (zh) 在trill网络中传输组播数据的方法、装置和系统
KR101145389B1 (ko) 분산화를 통한 확장성 있는 중앙 집중식 네트워크 구조 및 제어 방법과 이를 위한 네트워크 스위칭 장치
Bays et al. Reality shock in virtual network embedding: Flexibilizing demands for dealing with multiple operational requirements in SDNs
CN110300073A (zh) 级联端口的目标选择方法、聚合装置及存储介质

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: 19937704

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019937704

Country of ref document: EP

Effective date: 20220127