WO2019137355A1 - 传输数据的方法、设备和网络系统 - Google Patents

传输数据的方法、设备和网络系统 Download PDF

Info

Publication number
WO2019137355A1
WO2019137355A1 PCT/CN2019/070789 CN2019070789W WO2019137355A1 WO 2019137355 A1 WO2019137355 A1 WO 2019137355A1 CN 2019070789 W CN2019070789 W CN 2019070789W WO 2019137355 A1 WO2019137355 A1 WO 2019137355A1
Authority
WO
WIPO (PCT)
Prior art keywords
homed
user equipment
tunnel interface
data
access device
Prior art date
Application number
PCT/CN2019/070789
Other languages
English (en)
French (fr)
Inventor
李久勇
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 EP23158941.7A priority Critical patent/EP4221107A3/en
Priority to EP19738772.3A priority patent/EP3731472B1/en
Publication of WO2019137355A1 publication Critical patent/WO2019137355A1/zh
Priority to US16/924,734 priority patent/US11539551B2/en
Priority to US17/980,631 priority patent/US20230073291A1/en

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
    • H04L45/24—Multipath
    • H04L45/245—Link aggregation, e.g. trunking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4633—Interconnection of networks using encapsulation techniques, e.g. tunneling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • H04L12/4675—Dynamic sharing of VLAN information amongst network nodes
    • H04L12/4679—Arrangements for the registration or de-registration of VLAN attribute values, e.g. VLAN identifiers, port VLAN membership
    • 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
    • H04L45/24—Multipath
    • 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
    • H04L45/58—Association of routers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W60/00—Affiliation to network, e.g. registration; Terminating affiliation with the network, e.g. de-registration
    • H04W60/005—Multiple registrations, e.g. multihoming
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W88/00—Devices specially adapted for wireless communication networks, e.g. terminals, base stations or access point devices
    • H04W88/14—Backbone network devices
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L2101/00—Indexing scheme associated with group H04L61/00
    • H04L2101/60—Types of network addresses
    • H04L2101/677—Multiple interfaces, e.g. multihomed nodes
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

  • the present application relates to the field of communications and, more particularly, to a method, apparatus, and network system for transmitting data.
  • VXLAN Virtual eXtensible Local Area Network
  • VXLAN runs on existing network infrastructure and provides a way to extend Layer 2 networks.
  • VXLAN is a two-layer overlay scheme on a three-layer network. Each overlay network is referred to as a VXLAN network segment. Only virtual machines (VMs) in the same VXLAN network segment can communicate with each other.
  • Each VXLAN segment is identified by a 24-bit network segment identifier. The network segment identifier is called a VXLAN Network Identifier (VNI). This allows up to 16M VXLAN segments to exist simultaneously in a single management domain.
  • VNI VXLAN Network Identifier
  • the VNI is used to identify the range of inner MAC frames initiated by the VM. Therefore, there may be overlapping MAC addresses in the VXLAN network segment, but there is no cross traffic. Therefore the traffic is isolated by the VNI.
  • the VNI appears in the outer header of the inner MAC frame used to encapsulate the virtual machine.
  • VXLAN can also be called a tunneling scheme that superimposes a Layer 2 network on a Layer 3 network.
  • the tunnel is stateless, so each frame is encapsulated according to a set of rules.
  • the VXLAN Tunnel End Point (VTEP) is located on the Hypervisor of the server where the virtual machine is located. Therefore, the VNI and the VXLAN tunnel/outer package are only known by VTEP, but the VM never knows. It should be noted that VTEP can also be implemented on physical switches and physical servers, and can be implemented by software or hardware.
  • the present application provides a method, device, and network system for transmitting data, which implements traffic optimization of a single-homed user equipment, and reduces traffic pressure of the network system.
  • the first aspect provides a method for transmitting data, where the method is applied to a backbone device, where the backbone device is connected to at least two access devices, and the at least two access devices form an inter-device link aggregation group to connect multiple users.
  • the device, the at least one access device of the at least two access devices is further configured to connect to the single-homed user device, and each access device connected to the single-homed user device is configured with at least one single-homed tunnel interface, and the single-homed The tunnel interface is only used to transmit data that needs to be sent to the single-homed user equipment connected to the access device.
  • the backbone device stores a first mapping table, where the first mapping table is used to indicate each of the multiple single-homed user devices.
  • the tunnel interface identifier corresponding to the single-homed user equipment wherein the tunnel interface identifier corresponding to the single-homed user equipment is the identifier of the single-homed tunnel interface of the access device connected to the single-homed user equipment, and the method includes: Determining, according to the first mapping table, a first tunnel interface identifier corresponding to the first user equipment, when the first data that needs to be sent to the first user equipment is obtained, The first user equipment is a single-homed user equipment; the backbone device sends a first data packet including the first data to the first access device according to the first tunnel interface identifier, where the first access device is An access device configured with the identifier of the first tunnel interface.
  • the foregoing technical solution implements traffic optimization by setting at least one single-homed tunnel interface that is only used for transmitting data that needs to be sent to the single-homed user equipment on the access device connected to the single-homed user equipment, so that the traffic is optimized and sent to the single-homed user equipment.
  • Data does not circulate between multiple access devices to increase traffic pressure.
  • the packet is encapsulated according to different types of the port corresponding to the user equipment on the access device, so that the packets sent to different types of user equipment have different outer source IP addresses, so that the packets can be accurately selected in different VXLAN tunnels.
  • the single-homed tunnel interface forwards data so that the data is forwarded directly to the single-homed user equipment without Peer-link bypass, which reduces the bandwidth pressure of the Peer-link and optimizes the forwarding path of the single-homed user equipment.
  • each of the backbone device and the at least two access devices supports a virtual scalable local area network (VXLAN) function
  • the backbone device is configured according to the first a tunnel interface identifier
  • the first data packet including the first data is sent to the first access device, where the backbone device generates a first data packet, where the first data packet includes a VXLAN tunnel header, and the VXLAN tunnel head is external
  • the layer source address field includes the first tunnel interface identifier
  • the backbone device sends the first data packet to the first access device by using the first tunnel interface to identify the corresponding VXLAN tunnel.
  • inter-device link aggregation M-LAG access virtual scalable LAN VXLAN traffic optimization is implemented through the above technical solution, so that the main traffic in the VXLAN, that is, the unicast traffic destined for the single-homed user equipment does not pass the M-LAG.
  • the cross-link Peer-link forwarding between multiple member devices, that is, the access device reduces the pressure of the peer-link bandwidth and optimizes the forwarding path of the single-homed user equipment.
  • the first mapping table is specifically configured to indicate an identifier of each single-homed user equipment of the multiple single-homed user equipments Determining a mapping between the tunnel interface identifiers of the user equipment; determining the first tunnel interface identifier corresponding to the first user equipment according to the first mapping table, where the backbone device: according to the first mapping table, the first device The identifier of the corresponding tunnel interface is determined to be the first tunnel interface identifier, where the first device identifier is used to indicate the first user equipment.
  • the foregoing technical solution defines a first mapping table to indicate a mapping relationship between a device identifier and a tunnel interface identifier, and maps different tunnel interface identifiers by using different device identifiers of each user equipment, thereby implementing accurate selection of the backbone device.
  • a single-homed tunnel interface of the access device can forward the unicast packets sent to the single-homed user device to the access device through the single-homed tunnel interface.
  • the first data is data carried in a second data packet received by the backbone device, and, in the second data packet, The first device identification is included.
  • the packet received by the backbone device may be a unicast packet, a broadcast packet, a multicast packet, an unknown unicast packet, or the like (ie, a second data packet).
  • the first data may be sent to the single-homing user.
  • the data of the device is the data carried in the second data packet, and the second data packet includes the identifier of the single-homed user equipment.
  • the forwarding path of the unicast packet sent to the single-homed user equipment is also allowed to be forwarded directly to the single-homed user equipment without going through the horizontal link Peer-link bypass between the access devices. The pressure of the Peer-link bandwidth is reduced, and the forwarding path of the single-homed user equipment is optimized.
  • the backbone device obtains the first data by using a unicast manner.
  • the single-homed user equipment and the dual-homed user equipment are forwarded through a tunnel interface. If the destination message is sent to the single-homed user equipment, the single-homed user equipment is not connected. The access device needs to send the packet to the access device connected to the single-homed user equipment through the horizontal link Peer-link between the access devices, thus causing the horizontal link Peer-link Huge flow pressure.
  • the unicast traffic sent to the single-homed user equipment is differentiated from other traffic paths, and at least one data that needs to be sent to the single-homed user equipment is set for the access device connected to the single-homed user equipment.
  • the single-homed tunnel interface enables unicast traffic to be forwarded to the access device connected to the single-homed user equipment through the single-homed tunnel interface to optimize the traffic of the single-homed user equipment, so that the unicast traffic is sent to the single-homed user equipment. It does not forward through the horizontal link Peer-link between the M-LAG multiple member devices, that is, the access device, which reduces the pressure of the Peer-link bandwidth.
  • a second aspect provides a method for transmitting data, where the method is applied to a first access device, where the first access device and the second access device form an inter-device link aggregation group to connect the multi-homed user equipment.
  • the first access device is further configured to connect to the single-homed user equipment, where the first access device is configured with a single-homed tunnel interface, and the single-homed tunnel interface is only used for transmission and needs to be sent to the first access device.
  • the first access device receives the first data packet including the first data by using the first single-homed tunnel interface, where the first data is used as the first user equipment, where
  • the first single-homed tunnel interface is a single-homed tunnel interface indicated by the first tunnel interface identifier, and the first tunnel interface identifier is a tunnel interface identifier corresponding to the first user equipment, where the first user equipment is the first access device
  • the connected single-homed user equipment the first access device processes the first data packet to obtain the first data; and the first access device sends the first data to the first user equipment.
  • the first access device and the second access device support a virtual scalable local area network VXLAN function, where the first access device passes the first single-homed tunnel Receiving, by the interface, the first data packet that includes the first data, the first access device receiving the first data packet by using a VXLAN tunnel corresponding to the first single-homed tunnel interface, where the first data packet includes a VXLAN tunnel header, where The outer source address field of the VXLAN tunnel header includes the first tunnel interface identifier.
  • the first access device and the second access device are configured with a common dual-homing tunnel interface, where the dual-homing tunnel interface is used.
  • the data to be transmitted to the dual-homed user equipment is transmitted, and/or the dual-homing tunnel interface is used to transmit broadcast data, and/or the dual-homing tunnel interface is used for transmission to be sent to a single return that is not connected to the access device.
  • User device data is transmitted.
  • the original tunnel interface (ie, the dual-homing tunnel interface) of the access device is used to transmit data and broadcast data that need to be sent to the dual-homed user equipment, and needs to be sent to the non-connected device.
  • Single-homed at least one of the data of the user equipment, thereby achieving separation of the paths and reducing the pressure of the bandwidth.
  • a third aspect provides a backbone device, where the backbone device is configured to connect at least two access devices, and the at least two access devices form a cross-device link aggregation group to connect the multi-homed user device, where the at least two connections are At least one access device in the device is further configured to connect to the single-homed user device, and each access device connected to the single-homed user device is configured with at least one single-homed tunnel interface, and the single-homed tunnel interface is only used for transmission.
  • the backbone device comprising: a storage unit, configured to store a first mapping table, where the first mapping table is used to indicate each of the plurality of single-homed user devices
  • the tunnel interface identifier corresponding to the user equipment of the single-homed user equipment wherein the tunnel interface identifier corresponding to the single-homed user equipment is the identifier of the single-homed tunnel interface of the access device to which the single-homed user equipment is connected
  • the processing unit is configured to obtain the required Determining, by the first mapping table, a first tunnel interface identifier corresponding to the first user equipment, where the first user equipment is sent to the first user equipment, where the first user equipment is a single-homed user equipment
  • the communication unit is configured to send, according to the first tunnel interface identifier, a first data packet that includes the first data to the first access device, where the first access device is configured with the first
  • a tunnel interface identifies the access device of the single-homed tunnel interface indicated.
  • each of the backbone device and the at least two access devices supports a virtual scalable local area network VXLAN function
  • the processing unit is specifically configured to: Generating the first data packet, where the first data packet includes a VXLAN tunnel header, and an outer source address field of the VXLAN tunnel header includes the first tunnel interface identifier; the communication unit is specifically configured to identify, by using the first tunnel interface identifier The VXLAN tunnel sends the first data packet to the first access device.
  • the first mapping table is specifically configured to indicate an identifier of each single-homed user equipment in the single-homed user equipment and the single-homed
  • the mapping between the tunnel interface identifiers of the user equipment, the processing unit is configured to determine, according to the first mapping table, the tunnel interface identifier corresponding to the first device identifier, the first tunnel interface identifier, the first device identifier Used to indicate the first user equipment.
  • the first data is data carried in a second data packet received by the device, and the second data packet includes The first device identification.
  • the apparatus obtains the first data by using a unicast manner.
  • the fourth aspect provides an access device, where the access device is a first access device, and the first access device and the second access device form an inter-device link aggregation group to connect the multi-homed user device, where The first access device is further configured to connect to the single-homed user equipment, where the first access device is configured with a single-homed tunnel interface, and the single-homed tunnel interface is only used for transmitting a single that needs to be sent to the first access device.
  • the data of the user equipment, the first access device includes: a first communications unit, configured to receive, by using the first single-homed tunnel interface, a first data packet that includes the first data, where the first data is first
  • the first single-homed tunnel interface is a single-homed tunnel interface indicated by the first tunnel interface identifier, and the first tunnel interface identifier is a tunnel interface identifier corresponding to the first user equipment, where the first user equipment is a single-homed user equipment connected to the access device
  • a processing unit configured to process the first data packet to obtain the first data
  • a second communication unit configured to send the first data to the first user equipment.
  • the first access device and the second access device support a virtual scalable local area network VXLAN function, where the communication unit is specifically used by the first single return
  • the tunnel interface receives the first data packet, where the first data packet includes a VXLAN tunnel header, and an outer source address field of the VXLAN tunnel header includes the first tunnel interface identifier.
  • each device connected to the dual-homed user equipment is provided with at least one dual-homing port, and the dual-homing port is used for transmission.
  • the data sent to the dual-homed user equipment, and/or the dual-homed port is used to transmit broadcast data, and/or the dual-homed port is used to transmit data that needs to be sent to a single-homed user equipment that is not connected to the device.
  • At least one single-only port for transmitting data to be sent to the single-homed user equipment is set by the access device connected to the single-homed user equipment.
  • the unicast traffic of the single-homed user equipment is optimally forwarded, so that the unicast traffic sent to the single-homed user equipment in the VXLAN does not pass between the M-LAG multiple member devices.
  • the horizontal link Peer-link forwards the packets.
  • the packets are encapsulated according to the type of the port, so that the packets have different source IP addresses. Therefore, the single-homed tunnel interface can be accurately selected in different VXLAN tunnels to forward data.
  • the Peer-link bypass is directly forwarded to the single-homed user equipment, which reduces the bandwidth of the Peer-link and optimizes the forwarding path of the single-homed user equipment.
  • an apparatus for transmitting data may be a device for transmitting data in the above method design, or a chip configured in a device for transmitting data.
  • the device includes a processor, a memory, a communication interface, and a bus.
  • the processor, the memory, and the communication interface communicate through the bus.
  • the memory stores program code, and the processor can invoke the program code stored in the memory to perform the methods of transmitting data in the first and second aspects, and in any of the possible implementations of the first and second aspects.
  • the sixth aspect provides a network system, including at least one user equipment, a backbone device according to any one of the third aspects, and at least two access devices according to any one of the fourth aspect, wherein the backbone device is connected At least two access devices, the at least two access devices form a cross-device link aggregation group to connect the multi-homed user equipment, and at least one of the at least two access devices is further configured to connect to the single-homed user equipment
  • Each access device connected to the single-homed user equipment is provided with at least one single-homed tunnel interface, and the single-homed tunnel interface is only used to transmit data that needs to be sent to the single-homed user equipment connected to the access device.
  • a computer readable storage medium for storing computer instructions that, when executed on a computer, cause the computer to perform the first and second aspects and the first and the first A method of transmitting data in any of the possible implementations of the two aspects.
  • a computer program product comprising: computer program code, when the computer program code is run on a computer, causing the computer to perform the first and second aspects and the first and the first A method of transmitting data in any of the possible implementations of the two aspects.
  • FIG. 1 is a schematic diagram of a network system for accessing a VXLAN by using an M-LAG according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of packet forwarding of an M-LAG connected to a VXLAN according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of a configuration of an M-LAG accessing a VXLAN according to an embodiment of the present application
  • FIG. 4 is a schematic flowchart of a method for transmitting data in an embodiment of the present application.
  • FIG. 5 is still another schematic flowchart of a method for transmitting data in an embodiment of the present application.
  • FIG. 6 is a schematic block diagram of a backbone device in an embodiment of the present application.
  • FIG. 7 is a schematic block diagram of an access device in an embodiment of the present application.
  • FIG. 8 is a schematic block diagram of an apparatus 800 for transmitting data according to an embodiment of the present application.
  • FIG. 9 is a schematic block diagram of another apparatus 900 for transmitting data according to an embodiment of the present application.
  • Unicast means that the reception and delivery of information takes place only between two nodes. Communication between network nodes is like a conversation between people. If one person speaks to another person, then the terminology of network technology is used to describe "unicast". Unicast has a wide range of applications in the network, and most of the data on the network is transmitted in unicast form. Only the average network user does not know it. For example, when you send and receive emails and browse web pages, you must establish a connection with the mail server and web server. In this case, you use unicast data transmission. But “point to point” is often used instead of “unicast” because “unicast” generally corresponds to "multicast” and "broadcast”.
  • Multicast also known as "multicast" is not a lot of applications in network technology. Online video conferencing and online video on demand are particularly suitable for multicast. Because if unicast is used, it is transmitted on a node-by-node basis, and how many target nodes there are, how many transmission processes will occur. This method is obviously inefficient and undesirable; if a broadcast method that does not distinguish between targets and all transmissions is adopted, Although the data can be transmitted at one time, it is obviously not possible to distinguish the specific data receiving object. By means of multicast, it is possible to transfer data of all target nodes at a time, and to transfer data only to specific objects. Nodes that communicate with each other in multicast need to join the same multicast group. Multicasting of IP networks is typically accomplished by multicasting IP addresses. The multicast IP address is the class D IP address, which is the IP address between 224.0.0.0 and 239.255.255.255.
  • Broadcasting is widely used in the network. For example, the process of automatically obtaining an IP address by a client through a Dynamic Host Configuration Protocol (DHCP) is implemented by broadcasting. But compared to unicast and multicast, broadcasts take up almost all of the bandwidth of the network within the subnet. Similar to the meeting, only one person can speak at the meeting place. If all the people use the microphone to speak at the same time, the meeting will be a mess. Because the working principle of the hub determines that it is impossible to filter the broadcast storm, the general switch does not have this function. However, some network switches (such as the omnidirectional QS series switch) also have the function of filtering the broadcast storm. The router itself has isolated broadcast. The role of the storm.
  • DHCP Dynamic Host Configuration Protocol
  • the broadcast storm cannot be completely eliminated, but it can only be transmitted in the same subnet, just as the sound of the microphone can only be transmitted in the same venue. Therefore, in a large and medium-sized LAN composed of hundreds or even thousands of computers, it is generally carried out. Subnetting is like separating a hall from a wall into a number of small halls to isolate the broadcast storm.
  • the broadcast address is represented by the IP address "255.255.255.255", which represents all IP addresses in the same subnet.
  • BUM packets include broadcast packets, multicast packets, and unknown unicast packets.
  • FIG. 1 is an M-LAG access VXLAN of the present application. Schematic diagram of network system 100.
  • the VXLAN involved in this application is a typical two-layer network architecture.
  • the VXLAN includes a backbone layer and an access layer.
  • the backbone layer includes at least one backbone device, such as a backbone node (Spine), and the access layer includes a plurality of access devices, for example, leaf nodes.
  • the service gateway is generally deployed at the backbone layer, and the access layer is configured with at least one leaf node.
  • the user equipment accesses the VXLAN through the leaf node.
  • the backbone layer includes a backbone node Spine 101
  • the access layer includes two leaf nodes Leaf 102 and Leaf 103, and the Spine 101 and the access layer leaf nodes Leaf 102 and Leaf 103 are established.
  • Routing protocols communicate with each other.
  • Spine 101 can be directly connected to Leaf 102 and Leaf 103. It can also establish communication connections through other routing devices to complete tunnel establishment and host route synchronization.
  • User devices VM1, VM2, and VM3 shown at 105, 106, 107 access VXLAN through Leaf 102 and/or Leaf 103.
  • the user equipment can be connected to different leaf nodes through multiple different links, and the multiple links form an M-LAG, and each of the multiple links A link is called a member link of the M-LAG, and a device corresponding to the multiple links is called a member device of the M-LAG.
  • Each link has a corresponding port on the corresponding leaf node, and the corresponding port of the multiple links on different leaf nodes is called a member port of the M-LAG.
  • the multiple links in the M-LAG are used to load the traffic corresponding to the user equipment, and are mutually backup links. For example, in FIG. 1, VM2 is connected to Leaf 102 through Link1 shown in FIG.
  • Link1 and Link2 form M-LAG1, and member port 1 of M-LAG1 is located on Leaf 102, Port2.
  • Leaf 102 and Leaf 103 are member devices of M-LAG1, Link1 and Link2 are M-LAG1 member links, and Link1 and Link2 are used for load balancing of VM2 traffic and are backup links.
  • M-LAG technology can implement link aggregation between multiple devices, which improves link reliability from the board level to the device level.
  • a direct link must exist between the two devices that deploy the M-LAG.
  • the link must be configured as a horizontal link Peer-link.
  • Peer-link is a Layer 2 link used to negotiate packet exchange and part of traffic. After the interface on the leaf node is configured as a peer-link interface, no other services can be configured on the interface.
  • the user equipment When the user equipment is connected to only one access device, the user equipment is referred to as a single-homed user equipment, and the port used by the access device to connect the single-homed user equipment is a single-homed port, such as VM1 and VM3 in the network system 100. It is a single-homed user device, Port3 is a single-homed port of Leaf 102, and Port 4 is a single-homed port of Leaf 103.
  • the user equipment When a user equipment is connected to two access devices, the user equipment is called a dual-homed user equipment, and the access device is called The port connected to the dual-homed user device is a dual-homed port.
  • VM2 in the network system 100 is a dual-homed user device
  • Port 1 is a dual-homed port of Leaf 102
  • Port 2 is a dual-homed port of Leaf 103.
  • User equipment can also be connected to more than three access devices.
  • a dual-homed user equipment and a user equipment connected to three or more access devices are collectively referred to as a multi-homing access device.
  • the tunnel interface is the tunnel endpoint VTEP.
  • VTEP is deployed on Spine and Leaf respectively.
  • Each VTEP has a corresponding tunnel interface identifier, which may be an IP address.
  • the tunnel interface identifier is used to indicate the tunnel interface of the tunnel established between the Spine and the leaf.
  • the tunnel interface identifier corresponding to the single-homed user equipment is used to indicate the single-homed tunnel interface of the access device to which the single-homed user equipment is connected.
  • the single-homed tunnel interface is a tunnel outbound interface set in the access device connected to the single-homed user equipment and used for transmitting data that needs to be sent to the single-homed user equipment.
  • the VTEP of Spine 101 is VTEP1
  • the IP address of VTEP1 is IP1.
  • the VTEP of Leaf 102 and Leaf 103 is VTEP2, and the IP address of VTEP2 is IP2.
  • the VTEP shown in the figures is equivalent to a logical tunnel outbound interface, which is shown in the figure for ease of understanding.
  • the VXLAN packet is obtained.
  • the outer source IP address of the VXLAN packet is the local VTEP address
  • the outer destination IP address is the peer VTEP address
  • the pair of VTEP addresses correspond to a VXLAN tunnel.
  • the M-LAG multiple member devices need to be configured the same in the VXLAN.
  • VTEPIP address and MAC address For example, in Figure 1, Leaf 102 and Leaf 103 have the same tunnel interface VTEP2, i.e., have the same VTEP IP address IP2.
  • ECMP equal-cost multi-path
  • the traffic is distributed to the Leaf 102 and the Leaf 103 through the M-LAG load, and the leaf 102 and the Leaf 103 adopt the local preferential forwarding policy. Some of the received traffic is forwarded to the network side.
  • the user side and the network side are opposite to the access device, and the side where the access device is connected to the user device is called the user side, and the side where the access device is connected to the backbone device is called the network side.
  • the traffic sent from the network side to the user side is forwarded to the Leaf 103 or forwarded to the Leaf 103, and the two forwarding paths are not different.
  • the two forwarding paths are not different.
  • Leaf 102 determines that VM3 107 is not connected to Leaf 102, then Leaf 102 forwards the traffic to Leaf 103 via Peer-link 104, which will be This traffic is forwarded to VM3 107, and such a forwarding path causes unnecessary traffic to bypass.
  • FIG. 2 is a schematic diagram of packet forwarding of an M-LAG connected to a VXLAN according to the present application.
  • the message to be sent to the single-homed user equipment VM3 107 is sent via the tunnel interface VTEP1 of Spine 101, because Leaf 102 and Leaf 103 are configured with the same VTEP interface VTEP2, and have the same IP address IP2 of VTEP2. Therefore, the packet is forwarded to any one of the Leaf 102 and the Leaf 103. If the packet destined for the single-homed user equipment VM3 107 is forwarded to the Leaf 102, the Leaf 102 needs to forward the packet through the Peer-link 104.
  • the message can be forwarded normally to the single-homed user equipment VM3 107.
  • the message can be forwarded normally to the single-homed user equipment VM3 107.
  • Packe 101 sends a packet, it selects either Leaf 102 and Leaf 103. Therefore, for unicast packets sent to the single-homed device, there will be half the probability of passing through Leaf 102 and Leaf.
  • the Peer-link 104 between the 103s is bound to increase the bandwidth pressure of the Peer-link 104.
  • the Peer-link needs to bear the traffic generated by the BUM packets sent to the single-homed user equipment and the unicast packets, and is also connected to the M-LAG.
  • the proportion of BUM packets is very small, and most of them are unicast packets sent to the single-homed device. Therefore, most of the Peer-link bandwidth is sent to the single-homed user device. The broadcast message is occupied. Therefore, a large amount of unicast traffic sent from the network side to the user side of the single-homed port causes huge traffic pressure on the Peer-link.
  • An embodiment of the present application is directed to providing a traffic optimization scheme for a single-homed user equipment, where at least one single-homed tunnel that is only required to be transmitted to the single-homed user equipment data is set by the access device connected to the single-homed user equipment.
  • the interface is used to optimize the traffic of the M-LAG to the single-homed user equipment in the VXLAN.
  • the unicast traffic destined for the single-homed user equipment is not forwarded through the horizontal link Peer-link between the M-LAG member devices.
  • the packets sent to different types of user equipments have different outer source IP addresses according to different types of packets encapsulated by the user equipments on the access device, so that the packets can be accurately in different VXLAN tunnels.
  • the single-homed tunnel interface is selected to forward data, so that the data is directly forwarded to the single-homed user equipment without Peer-link bypassing, which reduces the bandwidth pressure of the Peer-link and optimizes the forwarding path of the single-homed user equipment.
  • FIG. 3 is a schematic diagram of a configuration of a network system based on M-LAG accessing a VXLAN according to an embodiment of the present application.
  • the network system 300 includes at least one user equipment (for example, VM1 and VM3 shown by the single-homed user equipments 105 and 107 shown in FIG. 3, VM2 shown by the dual-homed user equipment 106), backbone.
  • the device Spine 101 and at least two access devices (eg, Leaf 102 and Leaf 103), wherein one user device has a communication connection with at least one access device.
  • Each access device connected to the single-homed user device is provided with at least one single-homed port (for example, Port3 and Port4 in FIG.
  • the single-homed port Port3 is only used for transmission and needs to be sent to connect with the access device Leaf 102.
  • the data of VM1, Port4 is only used to transmit data that needs to be sent to VM3 107 connected to Leaf 103; and each access device connected to the dual-homed user equipment is provided with at least one dual-homing port, as shown in FIG. Port1 and Port2, the dual-homed port can be used to transmit data that needs to be sent to the dual-homed user equipment, or to transmit broadcast data, or for transmission to a single-homed user equipment that is not connected to the access device.
  • the data, in particular, the dual-homing port can be used to transmit a BUM message including a broadcast message, a multicast message, and an unknown unicast message.
  • the Leaf 102 in FIG. 3 is connected to the single-homed user equipment VM1 105, and the Leaf 103 is connected to the single-homed user equipment VM3 107, so VTEP3 is configured for Leaf 102 and VTEP4 is configured for Leaf 103.
  • VTEP3 is only used to transmit data that needs to be sent to the single-homed user equipment of Leaf 102, such as VM1 105.
  • VTEP4 is only used to transmit data that needs to be sent to the single-homed user equipment of Leaf 103, such as VM3 107.
  • the backbone device Spine 101 is configured with a conventional tunnel peer pointing to VTEP2, and two additional M-LAG single-homed peers are respectively directed to VTEP3 and VTEP4. Based on the above configuration, the Spine 101 and the access device Leaf 102 and Leaf are configured. Tunnels are established between 103, such as tunnels VTEP1-VTEP3, VTEP1-VTEP2, VTEP1-VTEP4.
  • the following describes the forwarding manner of the unicast traffic sent from the network side to the user side of the single-homed user equipment according to the embodiment of the present application with reference to FIG. 4 to FIG.
  • the load is shared to the member links of the M-LAG, and the member links of the M-LAG are reached through the common link, and then the member device forwards the received part of the traffic according to the local priority principle. Go to the network side and then forward it by the network side device. Therefore, in the embodiment of the present application, the packet forwarding path from the network side to the user side is mainly described.
  • FIG. 4 is a schematic flowchart of a method 400 for transmitting data in an embodiment of the present application.
  • the method 400 can be applied to a backbone device, where the backbone device is connected to at least two access devices, and the at least two access devices form an M-LAG to connect the multi-homed user device, where the at least two access devices are At least one access device is further configured to connect to the single-homed user equipment, and each access device connected to the single-homed user equipment is provided with at least one single-homed tunnel interface, and the single-homed tunnel interface is only used for transmission and needs to be sent to Data of a single-homed user device connected to the access device.
  • the method 400 includes the following.
  • the backbone device obtains the first data that needs to be sent to the first user equipment, and determines the first tunnel interface identifier corresponding to the first user equipment according to the first mapping table, where the first user equipment It is a single-homed user device.
  • the backbone device stores a first mapping table, where the first mapping table is used to indicate a tunnel interface identifier corresponding to each of the plurality of single-homed user devices, where a single-homed user The tunnel interface identifier corresponding to the device is an identifier of the single-homed tunnel interface of the access device to which the single-homed user equipment is connected.
  • the backbone device obtains a first mapping table, and when the backbone device receives the first data that needs to be sent to the single-homed user device, determining, according to the first mapping table, the single-homed The single-homed tunnel interface identifier corresponding to the user equipment. It should be understood that the invention is not limited thereto.
  • the first mapping table may be used to indicate the tunnel interface identifier corresponding to the single-homed user equipments VM1 and VM3.
  • the first mapping table is used to determine the single-homed user equipment VM1.
  • the first mapping table may be specifically used to indicate a mapping relationship between multiple device identifiers and multiple tunnel interface identifiers, where one device identifier can uniquely indicate one
  • the tunnel interface identifier is used to indicate the first user equipment.
  • a different tunnel identifier is mapped to each user device.
  • the backbone device forwards the data to the tunnel corresponding to the single-homed tunnel interface of the access device, so that the unicast packet sent to the single-homed user device is accurate. Forwarding to the access device accessed by the single-homed user equipment, and achieving the optimal forwarding of the unicast traffic of the single-homed port.
  • the access port Port3 of the single-homed user device VM1 is configured with a single-homed VTEP enable attribute
  • the access port Port4 of the VM3 is also configured with a single-homed VTEP enable attribute
  • the single-homed port enable of the single-homed user device VM1 is enabled.
  • the attribute corresponds to the tunnel interface identifier of the tunnel interface VTEP3
  • the single-homed port enable attribute of the single-homed user equipment VM3 corresponds to the tunnel interface identifier of the tunnel interface VTEP4
  • the medium access control MAC address of the single-homed user equipment VM1 corresponds to the tunnel interface.
  • the tunnel interface identifier of the VTEP3 and the medium access control MAC address of the single-homed user equipment VM3 correspond to the tunnel interface identifier of the tunnel interface VTEP4, so that the backbone device forwards the data to the single-homed tunnel interface of the access device, and sends it to the single-homed tunnel interface.
  • the unicast packet of the user equipment is accurately forwarded to the access device accessed by the single-homed user equipment, so that the unicast traffic of the single-homed user equipment is optimally forwarded.
  • the backbone device sends a first data packet including the first data to the first access device according to the first tunnel interface identifier, where the first access device is configured with the The access device identified by the first tunnel interface.
  • the backbone device Spine obtains the first data that needs to be sent to the VM1 (ie, the first user equipment)
  • the first tunnel interface identifier corresponding to the VM1 is determined according to the first mapping table, thereby determining the packet. The path forwarded.
  • the backbone device Spine determines, according to the first mapping table, a tunnel interface identifier corresponding to the device identifier of the user equipment, as the first tunnel interface identifier, for example, a VTEP3 tunnel corresponding to VM1 configured with a single-port port enable attribute.
  • the interface identifier is determined as the first tunnel interface identifier.
  • the VTEP1-VTEP3 tunnel is determined as the tunnel through which the packet forwarding path passes.
  • the medium access control MAC address of the VM1 corresponds to the tunnel interface identifier of the tunnel interface VTEP3.
  • the backbone device Spine determines the tunnel of VTEP1-VTEP3 as the tunnel through which the packet forwarding path passes.
  • the first data may be data in a second data packet that is received by the backbone device Spine and includes the first device identifier.
  • the second data may be a unicast packet, a broadcast packet, a multicast packet, or an unknown unicast packet.
  • the first data is a packet that needs to be sent to the single-homed user equipment VM1 in the second data packet.
  • a data includes the device identification of VM1.
  • the first device identification may also be a destination MAC address in the second data packet.
  • the single-homed tunnel interface VTEP can be accurately selected according to the destination MAC address, and the first data is encapsulated according to the unicast tunnel interface.
  • the first data after the encapsulation is forwarded by the tunnel corresponding to the single-homed tunnel interface, so that the forwarding path of the single-homed user equipment and the dual-homed user equipment is separated, the bandwidth of the Peer-link is reduced, and the single-homing user is optimized.
  • the forwarding path of the device is optimized.
  • the first data may be data carried by a unicast message.
  • the VXLAN encapsulation is performed on the tunnel interface VTEP1.
  • the format of the encapsulated VXLAN packet is as shown in Table 1.
  • the outer IP address of the encapsulated VXLAN packet carries the outer source IP address and the outer destination IP address.
  • the outer source IP address is the IP address of the VTEP of the Spine, and the outer destination IP address is used.
  • the VTEP IP address of the tunnel interface corresponding to the VM for example, the packet sent by Spine to VM1, the outer source IP address is IP address IP1 of VTEP1, and the outer destination IP address is the IP address IP3 of the single-homed tunnel interface VTEP3.
  • the outer destination IP address of the Outer IP header can be unicast and multicast.
  • the outer destination IP address is the IP address of the VTEP.
  • the VXLAN management layer is introduced, and the VXLAN Network Identifier (VNI) and the IP multicast group mapping are used to determine the VTEP. .
  • VNI VXLAN Network Identifier
  • the backbone device Spine sends a first data packet including the first data to the first access device by using the VXLAN according to the first tunnel interface identifier, where the first access device is configured with the single return indicated by the first tunnel interface identifier. Tunnel interface.
  • the backbone device Spine forwards the original packet sent to the single-homed user device VM1 to the VXLAN according to the tunnel interface identifier of the VTEP3, and forwards the obtained VXLAN packet to the leaf 102 through the tunnel identified by VTEP1-VTEP3.
  • the leaf 102 receives the VXLAN packet through the single-homed tunnel interface VTEP3, decapsulates the VXLAN packet, and obtains the original packet sent to the single-homed user device VM1, and then passes the original report through the single port port 3 of the Leaf 102 and VM1.
  • the text is sent to VM1.
  • the embodiment of the present invention implements traffic optimization of a single-homed access device that accesses the M-LAG to the virtual scalable local area network (VXLAN) across the device link aggregation network.
  • VXLAN virtual scalable local area network
  • the connection is connected through the single-homed user equipment.
  • the inbound device sets at least one single-homed tunnel interface for transmitting data to be sent to the single-homed user device, so that the main traffic in the VXLAN, that is, the unicast traffic destined for the single-homed user device does not pass through the M-LAG multiple
  • the horizontal link Peer-link between the member devices, that is, the access device, is forwarded, which reduces the pressure of the peer-link bandwidth and optimizes the forwarding path of the single-homed user device.
  • the encapsulation entries are sent to and from the backbone device and the access device.
  • the corresponding outer source IP address is the IP address of the VTEP on the peer end.
  • the corresponding destination IP address is the IP address of the local VTEP. .
  • the device After receiving a VXLAN packet, the device (which can be a backbone device or an access device) can hit the decapsulation table of the device according to the outer source IP address and the outer destination IP address of the VXLAN, indicating that the device needs to terminate the device. If the device is a transit device, the VXLAN packet needs to continue to forward the VXLAN packet to the downstream device.
  • the decapsulation table generated by each device is as shown in Table 2.
  • the VXLAN tunnel is a Layer 2 outbound port and therefore broadcasts on the VXLAN.
  • a VXLAN tunnel header is added to the domain to allow BUM packets including broadcast packets, multicast packets, and unknown unicast packets to be copied to the VXLAN tunnel in the broadcast domain.
  • the table used to send BUM messages is the headend copy list of VXLAN. In Figure 3, there are two peers on the Spine 101 that point to the Leaf 102. The M-LAG single-return peer, VTEP1-VTEP3, does not need to be added to the headend replication list.
  • the BUM traffic from Spine 101 to Leaf 102 will be duplicated.
  • the header copy list on the specific three devices is shown in Table 2.
  • the Spine 101 can prevent the Spine 101 from transmitting to the same access device on different VXLAN tunnels during the process of selecting the single-homed tunnel interface or the dual-homing tunnel interface. Multiple BUM messages.
  • Spine sends packets destined for a single-homed user device and a dual-homed user device through the same tunnel interface. If Spine sends a packet destined for a single-homed user device to another access device different from the access device of the single-homed user device, the other access device needs to pass the horizontal direction between the access devices.
  • the link Peer-link forwards the message to the access device of the single-homed user equipment, which causes a huge traffic pressure on the horizontal link Peer-link.
  • At least one single-homed tunnel interface that is only required to be transmitted to the single-homed user equipment data is set by the access device connected to the single-homed user equipment, so that the unicast traffic is forwarded to the single-homed tunnel interface.
  • the unicast traffic of the single-homed user equipment is optimally forwarded by the access device connected to the user equipment, so that the unicast traffic sent to the single-homed user equipment does not pass through the M-LAG multiple member devices.
  • the horizontal link Peer-link forwarding between the devices reduces the pressure of the peer-link bandwidth and optimizes the forwarding path of the single-homed user equipment.
  • the original tunnel interface of the access device (that is, the dual-homed tunnel interface) is used to transmit data that needs to be sent to the dual-homed user equipment, broadcast data, and data that needs to be sent to the single-homed user equipment that is not connected to the access device. At least one of the two, thereby achieving the separation of the path, reducing the pressure of the bandwidth.
  • FIG. 5 is still another schematic block diagram of a method 500 for transmitting data in an embodiment of the present application.
  • the method 500 can be applied to a first access device, where the first access device and the second access device form an M-LAG to connect to the multi-homed user device, and the first access device is further configured to connect to the single-homing device.
  • a user equipment the first access device is provided with a single-homed tunnel interface, and the single-homed tunnel interface is only used to transmit data that needs to be sent to the single-homed user equipment connected to the first access device, but
  • the application embodiment is not limited to this.
  • the method 500 includes the following.
  • the first access device receives the first data packet including the first data by using the first single-homing tunnel interface, where the first data is for the first user equipment, and the first single-homing
  • the tunnel interface is a single-homed tunnel interface indicated by the first tunnel interface identifier
  • the first tunnel interface identifier is a tunnel interface identifier corresponding to the first user equipment, where the first user equipment is the first access device Connected single-homed user devices.
  • S510 may be that the backbone device Spine 101 encapsulates the packet (first data) sent to the single-homed user device VM1 in a VXLAN packet (first data packet) and sends the packet to the Leaf 102.
  • the packet is forwarded to Leaf 102 through the tunnel identified by VTEP1-VTEP3, and Leaf 102 receives the VXLAN packet through VTEP3.
  • the Leaf 102 needs to receive a message to be sent to the user equipment VM1 connected to the Leaf 102 through the tunnel interface VTEP3. Then Leaf 102 first makes Spine101 learn VTEP3. When the leaf 102 forwards the packet sent by the VM1 to the Spine 101, it selects the outer source IP address according to whether the port of the VM1 is connected to the port of the VM1 (Port 3), and the single source VTEP is configured. The outer source IP address determined in the case of the energy is the IP address of VTEP3 (denoted as VTEP3), otherwise the determined outer source IP address is VTEP2. The processing on the Leaf 103 is similar and will not be described in detail.
  • the Spine 101 When receiving the packet of the VM1 sent by the Leaf 102, the Spine 101 learns that the outbound interface of the tunnel corresponding to the VM1 is VTEP3, and the Spine 101 receives the packet of the VM2 sent by the Leaf 102 or the Leaf 103, and learns that the outbound interface of the tunnel corresponding to the VM2 is VTEP2.
  • the Spine 101 receives the message of the VM3 sent by the Leaf 103, it learns that the tunnel outbound interface corresponding to the VM3 is VTEP4.
  • Table 3 the forwarding table learned on Spine 101 is shown in Table 3.
  • VM Forwarding entry VM1 VXLAN TUNNEL (VTEP3-VTEP1) VM2 VXLAN TUNNEL (VTEP2-VTEP1) VM3 VXLAN TUNNEL (VTEP4-VTEP1)
  • the unicast packets sent by the backbone device to the access device for example, the traffic sent from Spine 101 to Leaf 102
  • the outer destination IP addresses are VTEP2 and VTEP3.
  • the source IP address of the VXLAN header is encapsulated according to the type of the user equipment.
  • the forwarding path separation between the single-homed user equipment and the dual-homed user equipment can be achieved without separating the forwarding entries.
  • the BUM packets sent from the Spine 101 to the Leaf 102 and the Leaf 103 are different on the Spine 101, for example, broadcast packets, multicast packets, or unknown unicast packets sent from the backbone device to the access device.
  • the headend replication list in Table 2 Spine 101 will only copy one BUM message to VTEP2, and will not copy to VTEP3 and VTEP4.
  • the first access device processes the first data packet to obtain the first data.
  • the packet is encapsulated on the tunnel interface, and the first data packet received by the first access device includes the tunnel header of the VXLAN, so the first access device needs to decapsulate the first data packet to obtain The first data that needs to be sent to the first user equipment.
  • the first access device sends the first data to the first user equipment.
  • Leaf1 may send the first data to VM1.
  • the unicast traffic sent to the single-homed user equipment does not pass between the M-LAG multiple member devices.
  • the horizontal link Peer-link forwarding reduces the pressure of the peer-link bandwidth and optimizes the forwarding path of the unicast traffic to achieve optimal forwarding of unicast traffic of the single-homed user equipment.
  • the original port of the access device ie, dual-homed port
  • the dual-homed port is used to transmit broadcast data
  • the dual-homed port is used for The transmission needs to be sent to the data of the single-homed user equipment that is not connected to the access device, thereby realizing the separation of the paths and reducing the pressure of the bandwidth.
  • the solution provided by the embodiment of the present application implements traffic optimization of the single-homed access device of the M-LAG to access the VXLAN, and at least one of the access devices connected to the single-homed user device is only used for transmission and needs to be sent to the single-homing device.
  • the single-homed port of the user equipment data is used to optimize the traffic of the single-homed user equipment.
  • the unicast traffic sent to the single-homed user equipment does not pass through the M-LAG multiple member devices.
  • the horizontal link Peer-link is forwarded, and the packet is encapsulated according to the type of the port, so that the packet has different source IPs, so that the single-homed tunnel interface can be accurately selected in different VXLAN tunnels to forward data.
  • the data is directly forwarded to the single-homed user equipment without Peer-link bypassing, which reduces the pressure of the Peer-link bandwidth, optimizes the forwarding path of the single-homed port, and achieves the optimal forwarding of unicast traffic.
  • FIG. 6 is a schematic block diagram of a backbone device in an embodiment of the present application.
  • the backbone device 600 is configured in a network system including at least two access devices, where the at least two access devices have a communication connection for connecting the multi-homed user devices, and the at least two access devices At least one access device is further configured to connect to the single-homed user equipment, each single-homed user equipment is connected to only one access device, and each access device connected to the single-homed user equipment is provided with at least one for transmission only A single-homed tunnel interface that needs to be sent to the data of the single-homed user equipment connected to the access device.
  • the backbone device 600 may include: a storage unit 610, a processing unit 620, and a communication unit 630.
  • the storage unit 610 is configured to store a first mapping table, where the first mapping table is used to indicate a tunnel interface identifier corresponding to each of the plurality of single-homed user equipments, where a single-homed user equipment corresponds to
  • the tunnel interface identifier is a label of a single-homed tunnel interface of the access device to which the single-homed user equipment is connected.
  • the processing unit 620 is configured to determine, according to the first mapping table, a tunnel interface identifier corresponding to the single-homed user equipment when the communication unit 630 obtains the first data that needs to be sent to the single-homed user equipment.
  • the communication unit 630 is configured to send, according to the tunnel interface identifier determined by the processing unit 620, the single-homed tunnel interface indicated by the tunnel interface, to send, by the access device of the single-homed user equipment, first data including the first data. package.
  • the backbone device 600 may correspond to (for example, may be configured or be itself) the backbone device Spine described in the above method 400, and each module or unit in the device 600 is used to execute the backbone device in the method 400 described above, respectively.
  • the actions or processes performed by Spine may correspond to (for example, may be configured or be itself) the backbone device Spine described in the above method 400, and each module or unit in the device 600 is used to execute the backbone device in the method 400 described above, respectively. The actions or processes performed by Spine.
  • apparatus 600 for transmitting data shown in FIG. 6 is merely an example, and the apparatus for transmitting data in the embodiment of the present application may further include other modules or units, or include modules similar in function to the respective modules in FIG. Or not all of the modules in Figure 6 are included.
  • FIG. 7 is a schematic block diagram of an access device in an embodiment of the present application.
  • the access device is configured in a network system including a backbone device and at least two access devices, wherein at least two of the access devices have a communication connection between the multiple access user devices, and the at least two connections
  • At least one access device in the device is further configured to connect to the single-homed user device, each single-homed user device is connected to only one access device, and each access device connected to the single-homed user device is provided with at least one A single-homed tunnel interface for transmitting data that needs to be sent to a single-homed user equipment connected to the access device.
  • the access device 700 may include: a first communication unit 710, a processing unit 720, and a second communication unit 730.
  • the first communication unit 710 is specifically configured to receive, by using a single-homed tunnel interface, a first data packet that includes the first data, where the purpose of the first data is a single-homed user equipment, and the single-homed tunnel interface is a tunnel interface identifier.
  • the processing unit 720 is configured to process the first data packet to obtain the first data. As described above, in the VXLAN, the packet is encapsulated on the tunnel interface, and the first data packet received by the first communication unit 710 includes the tunnel header of the VXLAN. Therefore, the processing unit 720 needs to decapsulate the first data packet, and needs to send the packet. The first data to the first user device.
  • the second communication unit 730 is specifically configured to send the first data to the single-homed user equipment.
  • the access device 700 may correspond to (for example, may be configured or be itself) the access device described in the foregoing method 500, and each module or unit in the device 700 is used to perform the method 500 described above.
  • the detailed descriptions of the operations and processes performed by the device are omitted here to avoid redundancy.
  • apparatus 700 for transmitting data shown in FIG. 7 is merely an example, and the apparatus for transmitting data of the embodiment of the present application may further include other modules or units, or include modules similar in function to the respective modules in FIG. Or not all of the modules in Figure 7 are included.
  • FIG. 8 is a schematic block diagram of an apparatus 800 for transmitting data according to an embodiment of the present application.
  • the device may be a backbone device or an access device in the above method, and the device 800 includes a processor 810, a memory 820, a communication interface 830, and a bus 840.
  • the processor 810, the memory 820, and the communication interface 830 communicate via the bus 840, and may also communicate by other means such as wireless transmission.
  • the memory 820 is used to store instructions
  • the processor 810 is configured to execute instructions stored in the memory 820
  • the memory 820 stores program codes
  • the processor 810 can call program codes stored in the memory 820 to control the communication interface 830 to send and receive information.
  • the memory 820 further stores a first mapping table, where the first mapping table is used to indicate a tunnel interface identifier corresponding to each of the plurality of single-homed user devices, where The tunnel interface identifier corresponding to the single-homed user equipment is an identifier of the single-homed tunnel interface of the access device to which the single-homed user equipment is connected.
  • the processor 810 is configured to determine, according to the first mapping table, a first tunnel interface identifier corresponding to the first user equipment, when the first data that needs to be sent to the first user equipment is obtained, where the first user equipment is a single-homed user equipment; the processor 810 can also control the communication interface 830 to send, according to the first tunnel interface identifier, a first data packet including the first data to the first access device, where the first The access device is an access device configured with the single-homed port indicated by the first tunnel interface identifier.
  • the communication interface 830 can be configured to receive a first data packet including the first data through a single-homed tunnel interface (eg, corresponding to the Ethernet interface 914 in the network device 900) by the processor
  • the first data packet is processed by the 810, and the first data is obtained, and the first data is sent to the single-homed user equipment by using the communication interface 830, where the purpose of the first data is a single-homed user equipment
  • the single-homed tunnel interface is a single-homed tunnel interface indicated by the tunnel interface identifier
  • the tunnel interface identifier is a tunnel interface identifier corresponding to the single-homed user equipment.
  • FIG. 9 is a schematic block diagram of another apparatus 900 for transmitting data according to an embodiment of the present application.
  • the device 900 for example, a backbone device or an access device
  • the device 900 generally includes an interface board 910 and a main control board 920.
  • the interface board 910 and the main control board 920 can be connected through an internal bus 930.
  • the interface board 910 includes a central processing unit (CPU) 911, a memory 912, an Ethernet switch (LSW) 913, and an Ethernet interface 914.
  • the main control board 920 includes a CPU 921, a memory 922, and the like.
  • the memory of the main control board system and the board memory of the interface board are commonly mapped in the memory space of the network device, and the CPU is a control unit of the network device main control board and the interface board, and the memory may include, for example, a random access memory (random access memory, RAM) and/or flash memory, etc., running programs and statically configured parameters can be saved in flash memory, and code and data executed while the program is running can be placed in RAM connected to the CPU.
  • the LSW 913 can be configured with an external RAM to store the forwarded packet content, which can solve some scenarios in which the internal buffer of the LSW chip is insufficient.
  • the physical layer PHY connected to the LSW 913 completes the optical interface or electrical interface between the devices.
  • the Ethernet interface is docked.
  • the Ethernet switch chip LSW 913 of the interface board 910 of the network device receives the Ethernet data packet from the outside of the network device through the Ethernet interface 914, and the LSW 913 receives the CPU configuration command, forwarding entry, and the like. Guides the forwarding behavior of LSW 913 packets.
  • the CPU 921 and the CPU 911 may correspond to the processor 810 in the device 800, and the LSW 913 and the Ethernet interface 914 may correspond to the communication interface 830 in the device 800 described above.
  • the specific functions of each module or unit have been described in detail above, and are not described herein again for the sake of brevity.
  • the cooperation between the main control board and the CPU in the interface board cooperates to control the forwarding behavior of the LSW, and the traffic optimization of the single-homed user equipment of the M-LAG to the VXLAN is implemented, and the main traffic in the VXLAN can be made. That is, the unicast traffic sent to the single-homed user equipment will not be forwarded through the peer-link, which reduces the pressure on the bandwidth of the Peer-link.
  • the above method embodiments may be applied to a processor or implemented by a processor.
  • the processor may be an integrated circuit chip with signal processing capabilities.
  • each step of the foregoing method embodiment may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
  • the above processor may be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or the like.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • Programming logic devices, discrete gates or transistor logic devices, discrete hardware components The methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method.
  • the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory may be a read-only memory (ROM), a programmable read only memory (ROMM), an erasable programmable read only memory (erasable PROM, EPROM), or an electrical Erase programmable EPROM (EEPROM) or flash memory.
  • the volatile memory can be a random access memory (RAM) that acts as an external cache.
  • RAM random access memory
  • RAM random access memory
  • many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (Synchronous DRAM).
  • SDRAM double data rate synchronous DRAM
  • DDR SDRAM double data rate synchronous DRAM
  • ESDRAM enhanced synchronous dynamic random access memory
  • SLDRAM synchronously connected dynamic random access memory
  • DR RAM direct memory bus random access memory
  • the size of the sequence numbers of the foregoing processes does not mean the order of execution sequence, and the order of execution of each process should be determined by its function and internal logic, and the present invention should not be The implementation of the embodiments constitutes any limitation.
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read only memory, a random access memory, a magnetic disk, or an optical disk.

Landscapes

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

Abstract

本申请提供了一种传输数据的方法、设备和网络系统，该方法应用于骨干设备中，该骨干设备连接至少两个接入设备，该接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置单归隧道接口，该骨干设备存储有第一映射表，用于指示每个单归用户设备对应的隧道接口标识，该方法包括：当骨干设备获得需要发送给第一用户设备的第一数据时，根据第一映射表，确定第一用户设备对应的第一隧道接口标识，其中，第一用户设备为单归用户设备；该骨干设备根据第一隧道接口标识，向配置有该第一隧道接口标识的第一接入设备发送包括该第一数据的第一数据包，从而能够优化数据转发路径，实现单归用户设备的流量优化，降低该网络系统的流量压力。

Description

传输数据的方法、设备和网络系统
本申请要求于2018年1月11日提交中国专利局、申请号为201810026227.1、发明名称为“传输数据的方法、设备和网络系统”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
本申请涉及通信领域，并且更具体地，涉及一种传输数据的方法、设备和网络系统。
背景技术
虚拟可扩展局域网(Virtual eXtensible Local Area Network，VXLAN)用于解决多租户环境下的虚拟机对二层和三层数据中心网络基础设施的要求。VXLAN运行在现有的网络基础设施上，并提供了一种扩展二层网络的方式。总之，VXLAN是三层网络上的二层叠加方案。每个叠加网被称为一个VXLAN网段。只有同一个VXLAN网段里的虚拟机(virtual machine，VM)才能相互通信。每个VXLAN网段由一个24比特的网段标识识别。该网段标识被称为VXLAN标识(VXLAN Network Identifier，VNI)。这使得一个管理域内最多可以同时存在16M个VXLAN网段。
VNI用于识别由VM发起的内层MAC帧的范围。因此，VXLAN网段里可能有重叠的MAC地址，但是没有交叉的流量。因此流量被通过VNI隔离了。VNI出现在用于封装虚拟机发起的内层MAC帧的外层头中。
因为该封装，VXLAN也可以被称为在三层网络上叠加二层网络的隧道方案。该隧道是无状态的，因此，每个帧是根据一系列规则封装的。VXLAN隧道端点(VXLAN Tunnel End Point，VTEP)位于该虚拟机所在的服务器的Hypervisor上。因此，该VNI和该VXLAN隧道/外层封装仅仅被VTEP所知，但是VM从来不知道。需要注意的是，VTEP还可以在物理交换机和物理服务器上，并且可以通过软件或硬件实现。
发明内容
本申请提供一种传输数据的方法、设备和网络系统，实现单归用户设备的流量优化，降低该网络系统的流量压力。
第一方面，提供一种传输数据的方法，该方法应用于骨干设备中，该骨干设备连接至少两个接入设备，该至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，该至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，该单归隧道接口仅用于传输需要发送给与该接入设备连接的单归用户设备的数据，该骨干设备存储有第一映射表，该第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是该单归用户设备所连接的接入设备的单归隧道接口的标识，该方法包括：当骨干设备获得需要发送给第一用户设备的第一数据时，根据该第一映射表，确定该第一用户设备对应的第一隧道接口标识，其中，该第一用户 设备为单归用户设备；该骨干设备根据该第一隧道接口标识，向第一接入设备发送包括该第一数据的第一数据包，其中，该第一接入设备是配置有该第一隧道接口标识的接入设备。
上述技术方案通过在单归用户设备所连接的接入设备上设置至少一个仅用于传输需要发送给该单归用户设备数据的单归隧道接口，实现流量优化，使发往单归用户设备的数据不会在多个接入设备之间绕行增加流量压力。本实施例根据接入设备上用户设备对应的端口的不同类型封装报文，使发往不同类型的用户设备的报文具有不同的外层源IP地址，从而能够在不同的VXLAN隧道中精确选择单归隧道接口来转发数据，使数据不经过Peer-link绕行直接转发至该单归用户设备，降低了Peer-link的带宽压力，优化了单归用户设备的转发路径。
结合第一方面，在第一方面的第一种实现方式中，该骨干设备和该至少两个接入设备中的每个接入设备支持虚拟可扩展局域网络VXLAN功能，该骨干设备根据该第一隧道接口标识，向第一接入设备发送包括该第一数据的第一数据包，包括：该骨干设备生成第一数据包，该第一数据包包括VXLAN隧道头，该VXLAN隧道头的外层源地址字段包括第一隧道接口标识；该骨干设备通过所述第一隧道接口标识对应的VXLAN隧道向第一接入设备发送该第一数据包。
在跨设备链路聚合M-LAG接入虚拟可扩展局域网VXLAN中，通过上述技术方案实现流量优化，使VXLAN中的主要流量，即发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link转发，降低了Peer-link的带宽的压力，优化了单归用户设备的转发路径。
结合第一方面及其上述实现方式，在第一方面的第二种实现方式中，该第一映射表具体用于指示多个单归用户设备中每个单归用户设备的标识与该单归用户设备的隧道接口标识之间的映射关系；该根据该第一映射表，确定该第一用户设备对应的第一隧道接口标识，包括：该骨干设备根据该第一映射表，将第一设备标识对应的隧道接口标识，确定为该第一隧道接口标识，该第一设备标识用于指示该第一用户设备。
上述技术方案将第一映射表定义为用于指示设备标识和隧道接口标识之间的映射关系，通过每一台用户设备的不同的设备标识来映射不同的隧道接口标识，从而实现骨干设备精确选择接入设备的单归隧道接口，能够将发往单归用户设备的单播报文准确的通过单归隧道接口转发至接入设备，实现单归端口的单播流量最优转发的目的。
结合第一方面及其上述实现方式，在第一方面的第三种实现方式中，该第一数据是该骨干设备接收到的第二数据包中携带的数据，并且，该第二数据包中包括该第一设备标识。
骨干设备接收到的报文，可以是单播报文、广播报文、组播报文、未知单播报文等(即，第二数据包)，此时第一数据可以指需要发送到单归用户设备的数据，是第二数据包中携带的数据，第二数据包中包括单归用户设备的标识。在这种情况下，限定发往单归用户设备的单播报文的转发路径，也能够使数据不经过接入设备之间的横向链路Peer-link绕行直接转发至该单归用户设备，降低了Peer-link的带宽的压力，优化了单归用户设备的转发路径。
结合第一方面及其上述实现方式，在第一方面的第四种实现方式中，该骨干设备通 过单播方式获得该第一数据。
在现有技术中，发往单归用户设备和双归用户设备的不同类型的数据都是通过一个隧道接口转发，如果目的为单归用户设备的报文被发送给没有连接该单归用户设备的接入设备，该接入设备就需要通过接入设备之间的横向链路Peer-link将该报文发送给连接该单归用户设备的接入设备，所以导致横向链路Peer-link的巨大流量压力。本申请实施例将发往单归用户设备的单播流量和其他流量路径进行区分，通过为单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归隧道接口，使单播流量通过该单归隧道接口转发到该单归用户设备所连接的接入设备来实现单归用户设备的流量优化，使发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link进行转发，降低了Peer-link的带宽的压力。
第二方面，提供了一种传输数据的方法，该方法应用于第一接入设备中，该第一接入设备与第二接入设备组成跨设备链路聚合组以连接多归用户设备，该第一接入设备还用于连接单归用户设备，该第一接入设备中设置有单归隧道接口，该单归隧道接口仅用于传输需要发送给与该第一接入设备连接的单归用户设备的数据，该方法包括：该第一接入设备通过第一单归隧道接口接收包括第一数据的第一数据包，其中，该第一数据的目的为第一用户设备，该第一单归隧道接口是第一隧道接口标识指示的单归隧道接口，该第一隧道接口标识是该第一用户设备对应的隧道接口标识，该第一用户设备为与该第一接入设备连接的单归用户设备；该第一接入设备处理该第一数据包得到该第一数据；该第一接入设备向该第一用户设备发送该第一数据。
结合第二方面，在第二方面的第一种实现方式中，该第一接入设备和第二接入设备支持虚拟可扩展局域网络VXLAN功能，该第一接入设备通过第一单归隧道接口接收包括第一数据的第一数据包，包括：该第一接入设备通过该第一单归隧道接口对应的VXLAN隧道接收该第一数据包，该第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括该第一隧道接口标识。
结合第二方面及其上述实现方式，在第二方面的第二种实现方式中，该第一接入设备和第二接入设备中设置有共同的双归隧道接口，该双归隧道接口用于传输需要发送给双归用户设备的数据，和/或该双归隧道接口用于传输广播数据，和/或该双归隧道接口用于传输需要发送给不与该接入设备连接的单归用户设备的数据。
在上述技术方案中，接入设备原有的隧道接口(即双归隧道接口)，用来传输需要发送给双归用户设备的数据、广播数据，和需要发送给不与该接入设备连接的单归用户设备的数据中的至少一种，从而实现路径的分离，降低了带宽的压力。
第三方面，提供了一种骨干设备，该骨干设备用于连接至少两个接入设备，该至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，该至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，该单归隧道接口仅用于传输需要发送给与该接入设备连接的单归用户设备的数据，该骨干设备包括：存储单元，用于存储第一映射表，该第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是该单归用户设备所连接的接入设备的单归隧道接 口的标识；处理单元，用于在获得需要发送给第一用户设备的第一数据时，根据该第一映射表，确定该第一用户设备对应的第一隧道接口标识，该第一用户设备为单归用户设备；通信单元，用于根据该第一隧道接口标识，向第一接入设备发送包括该第一数据的第一数据包，其中，该第一接入设备是配置有该第一隧道接口标识所指示的单归隧道接口的接入设备。
结合第三方面，在第三方面的第一种实现方式中，该骨干设备和该至少两个接入设备中的每个接入设备支持虚拟可扩展局域网络VXLAN功能，该处理单元具体用于生成该第一数据包，该第一数据包包括VXLAN隧道头，该VXLAN隧道头的外层源地址字段包括该第一隧道接口标识；该通信单元具体用于通过该第一隧道接口标识对应的VXLAN隧道向该第一接入设备发送该第一数据包。
结合第三方面及其上述实现方式，在第三方面的第二种实现方式中，该第一映射表具体用于指示多个单归用户设备中每个单归用户设备的标识与该单归用户设备的隧道接口标识之间的映射关系，该处理单元具体用于根据该第一映射表，将第一设备标识对应的隧道接口标识，确定为该第一隧道接口标识，该第一设备标识用于指示该第一用户设备。
结合第三方面及其上述实现方式，在第三方面的第三种实现方式中，该第一数据是该装置接收到的第二数据包中携带的数据，并且，该第二数据包中包括该第一设备标识。
结合第三方面及其上述实现方式，在第三方面的第四种实现方式中，该装置通过单播方式获得该第一数据。
第四方面，提供了一种接入设备，该接入设备为第一接入设备，该第一接入设备与第二接入设备组成跨设备链路聚合组以连接多归用户设备，该第一接入设备还用于连接单归用户设备，该第一接入设备中设置有单归隧道接口，该单归隧道接口仅用于传输需要发送给与该第一接入设备连接的单归用户设备的数据，该第一接入设备包括：第一通信单元，用于通过第一单归隧道接口接收包括第一数据的第一数据包，其中，该第一数据的目的为第一用户设备，该第一单归隧道接口是第一隧道接口标识指示的单归隧道接口，该第一隧道接口标识是该第一用户设备对应的隧道接口标识，该第一用户设备为与该第一接入设备连接的单归用户设备；处理单元，用于处理该第一数据包，得到该第一数据；第二通信单元，用于向该第一用户设备发送该第一数据。
结合第四方面，在第四方面的第一种实现方式中，该第一接入设备和第二接入设备支持虚拟可扩展局域网络VXLAN功能，该通信单元具体用于由该第一单归隧道接口接收该第一数据包，该第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括该第一隧道接口标识。
结合第四方面及其上述实现方式，在第四方面的第二种实现方式中，每个连接有双归用户设备的该装置中设置有至少一个双归端口，该双归端口用于传输需要发送给双归用户设备的数据，和/或该双归端口用于传输广播数据，和/或该双归端口用于传输需要发送给不与该装置连接的单归用户设备的数据。
在跨设备链路聚合M-LAG接入虚拟可扩展局域网VXLAN中，通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归端口，实现单归用户设备的单播流量的最优转发，使VXLAN中的主要流量，发往单归用户设备 的单播流量不会通过M-LAG多台成员设备之间即该装置之间的横向链路Peer-link进行转发，而是根据端口的类型封装报文，使报文具有不同的源IP，从而能够实现在不同的VXLAN隧道中精确选择单归隧道接口来转发数据，使数据不经过Peer-link绕行直接转发至该单归用户设备，降低了Peer-link的带宽的压力优化了单归用户设备的转发路径。
第五方面，提供了一种传输数据的设备，所述设备可以为上述方法设计中传输数据的装置，或者为配置于传输数据的装置中的芯片。所述设备：包括处理器、存储器、通信接口和总线。其中，处理器、存储器、通信接口通过总线进行通信。该存储器存储程序代码，且处理器可以调用存储器中存储的程序代码执行第一和第二方面以及第一和第二方面中任一种可能实现方式中的传输数据的方法。
第六方面，提供了一种网络系统，包括至少一个用户设备、如第三方面的任一项的骨干设备和至少两个如第四方面任一项的接入设备，其中，该骨干设备连接至少两个接入设备，该至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，该至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，该单归隧道接口仅用于传输需要发送给与该接入设备连接的单归用户设备的数据。
第七方面，提供了一种计算机可读存储介质，其特征在于，用于存储计算机指令，当该计算机指令在计算机上运行时，使得该计算机执行上述第一和第二方面以及第一和第二方面中任一种可能实现方式中的传输数据的方法。
第八方面，提供了一种计算机程序产品，所述计算机程序产品包括：计算机程序代码，当所述计算机程序代码在计算机上运行时，使得计算机执行上述第一和第二方面以及第一和第二方面中任一种可能实现方式中的传输数据的方法。
上述第二至第八方面及其各实现方式的有益效果可以参考第一方面及各其实现方式的方法的有益效果。
附图说明
图1是本申请实施例的M-LAG接入VXLAN的网络系统示意图；
图2是本申请实施例的M-LAG接入VXLAN的一种报文转发示意图；
图3是本申请实施例的M-LAG接入VXLAN的一种配置示意图；
图4是本申请实施例中传输数据的方法的示意性流程图；
图5是本申请实施例中传输数据的方法的又一示意性流程图；
图6是本申请实施例中骨干设备的示意性框图；
图7是本申请实施例中接入设备的示意性框图；
图8是本申请实施例提供的一种传输数据的设备800的示意性框图；
图9是本申请实施例提供的另一种传输数据的设备900的示意性框图。
具体实施方式
为了便于理解方案，首先对本申请实施例中涉及到的一些概念作简单介绍。
单播，是指信息的接收和传递只在两个节点之间进行。网络节点之间的通信就好像是人们之间的对话一样。如果一个人对另外一个人说话，那么用网络技术的术语来描述 就是“单播”，单播在网络中有广泛的应用，网络上绝大部分的数据都是以单播的形式传输的，只是一般网络用户不知道而已。例如，你在收发电子邮件、浏览网页时，必须与邮件服务器、Web服务器建立连接，此时使用的就是单播数据传输方式。但是通常使用“点对点通信”(Point to Point)代替“单播”，因为“单播”一般与“多播”和“广播”相对应使用。
多播，也可以称为“组播”，在网络技术的应用并不是很多，网上视频会议、网上视频点播特别适合采用多播方式。因为如果采用单播方式，逐个节点传输，有多少个目标节点，就会有多少次传送过程，这种方式显然效率极低，是不可取的；如果采用不区分目标、全部发送的广播方式，虽然一次可以传送完数据，但是显然达不到区分特定数据接收对象的目的。采用多播方式，既可以实现一次传送所有目标节点的数据，也可以达到只对特定对象传送数据的目的。多播中相互通信的节点需要加入到同一个多播组中。IP网络的多播一般通过多播IP地址来实现。多播IP地址就是D类IP地址，即224.0.0.0至239.255.255.255之间的IP地址。
广播，在网络中的应用较多，如客户机通过动态主机配置协议(Dynamic host configuration protocol，DHCP)自动获得IP地址的过程就是通过广播来实现的。但是同单播和多播相比，广播几乎占用了子网内网络的所有带宽。类似于开会，在会场上只能有一个人发言，如果所有的人同时都用麦克风发言，会场上就会乱成一锅粥。集线器由于其工作原理决定了不可能过滤广播风暴，一般的交换机也没有这一功能，不过现在有的网络交换机(如全向的QS系列交换机)也有过滤广播风暴功能了，路由器本身就有隔离广播风暴的作用。广播风暴不能完全杜绝，但是只能在同一子网内传播，就好像麦克风的声音只能在同一会场内传播一样，因此在由几百台甚至上千台电脑构成的大中型局域网中，一般进行子网划分，就像将一个大厅用墙壁隔离成许多小厅一样，以达到隔离广播风暴的目的。在IP网络中，广播地址用IP地址“255.255.255.255”来表示，这个IP地址代表同一子网内所有的IP地址。
BUM报文包括广播报文、组播报文以及未知单播报文。
下面将结合附图，对本申请中的技术方案进行描述。
下面将以跨设备的链路聚合(multi-chassis link aggregation，M-LAG)接入虚拟可扩展局域网VXLAN的网络系统为例详细描述本申请实施例，图1是本申请M-LAG接入VXLAN的网络系统100示意图。
本申请所涉及的VXLAN是一种典型的两层网络架构。VXLAN包括骨干层和接入层，骨干层包括至少一个骨干设备，例如骨干节点(Spine),接入层包括多个接入设备，例如，叶子节点(Leaf)。业务网关一般部署在骨干层，接入层设置至少一个叶子节点。用户设备通过叶子节点接入VXLAN。如图1所示，本申请中以骨干层包括一个骨干节点Spine 101，接入层包括两个叶子节点Leaf 102和Leaf 103，Spine 101和接入层的叶子节点Leaf 102和Leaf 103之间建立路由协议以相互通信。Spine 101与Leaf 102和Leaf 103可以直接连接，也可以通过其他路由设备建立通信连接，完成隧道的建立和主机路由同步。105、106、107所示的用户设备VM1、VM2和VM3通过Leaf 102和/或Leaf 103接入VXLAN。
为了为用户设备提供接入的可靠性，可以使用户设备通过多条不同的链路接入到不 同的叶子节点上，该多条链路组成M-LAG，该多条链路中的每条链路称为该M-LAG的成员链路，该多条链路对应的设备称为该M-LAG的成员设备。每条链路在对应的叶子节点上都有对应的端口，将该多条链路在不同叶子节点上对应的端口称为该M-LAG的成员端口。该M-LAG中的多条链路用于负载分担用户设备对应的流量，并且互为备份链路。例如，图1中，VM2通过108所示的Link1连接Leaf 102，通过109所示的Link2连接Leaf 103，Link1和Link2组成了M-LAG1，该M-LAG1的成员端口Port1位于Leaf 102上，Port2位于Leaf 103上，则Leaf 102和Leaf 103为M-LAG1的成员设备，Link1和Link2为M-LAG1成员链路，Link1和Link2用于实现VM2的流量的负载分担，并且互为备份链路。
M-LAG技术能够实现多台设备间的链路聚合，从而把链路可靠性从单板级提高到了设备级。部署M-LAG的两台设备之间必须存在的一条直连链路，且该链路必须配置为横向链路Peer-link。Peer-link是一条二层链路，用于协商报文的交互及部分流量的传输。叶子节点上的接口被配置为Peer-link接口后，该接口上不能再配置其它业务。
当用户设备仅仅连接到一个接入设备时，称该用户设备为单归用户设备，该接入设备用于连接该单归用户设备的端口为单归端口，例如网络系统100中的VM1和VM3就为单归用户设备，Port3为Leaf 102的单归端口，Port4为Leaf 103的单归端口；当用户设备连接到2个接入设备时，称该用户设备为双归用户设备，接入设备连接该双归用户设备的端口为双归端口，例如网络系统100中的VM2就为双归用户设备，Port1为Leaf 102的双归端口，Port2为Leaf 103的双归端口。用户设备还可以连接到三个以上的接入设备。本申请中，将双归用户设备和连接到三个以上的接入设备的用户设备统称为多归接入设备。
在M-LAG接入VXLAN的网络系统中，隧道接口即为隧道端点VTEP。VTEP分别部署在Spine和Leaf上。每个VTEP具有对应的隧道接口标识，该隧道接口标识可以为IP地址。该隧道接口标识用于指示Spine和Leaf之间建立的隧道的隧道接口，例如，一个单归用户设备对应的隧道接口标识用于指示该单归用户设备所连接的接入设备的单归隧道接口，单归隧道接口是连接有单归用户设备的接入设备中设置的仅用于传输需要发送给该单归用户设备的数据的隧道出接口。
图1中，Spine 101的VTEP为VTEP1，VTEP1的IP地址为IP1。Leaf 102和Leaf 103的VTEP为VTEP2，VTEP2的IP地址为IP2。在本申请中，附图中所示的VTEP等同于一个逻辑上的隧道出接口，为便于理解，在图中示出。报文经过VTEP封装后得到VXLAN报文，VXLAN报文中的外层源IP地址为本端VTEP地址，外层目的IP地址为对端VTEP地址，一对VTEP地址就对应着一个VXLAN隧道。
由于M-LAG的多台成员设备对上层设备体现的是一台设备，即M-LAG多台成员设备会虚拟化成一台逻辑设备，因此在VXLAN中M-LAG的多台成员设备需要配置相同的VTEPIP地址和MAC地址。例如在图1中，Leaf 102和Leaf 103具有相同的隧道接口VTEP2，即具有相同的VTEP IP地址IP2。当网络侧向用户侧转发图1中黑实线所示的由北向南的流量时，该流量会通过等价多路径(equal-cost multi-path，ECMP)负载分担到Leaf 102和Leaf 103上。而用户侧向网络侧发送图1中虚线所示的由南向北的流量时，该流量是通过M-LAG负载分担到Leaf 102和Leaf 103，Leaf 102和Leaf 103采用本地优 先转发的策略将接收到的部分流量转发到网络侧。本申请实施例中，用户侧和网络侧是相对接入设备来说的，将接入设备连接用户设备的一侧称为用户侧，将接入设备连接骨干设备的一侧称为网络侧。
对于双归用户设备VM2 106来说，从网络侧发往用户侧的流量不论转发至Leaf 102，还是转发至Leaf 103，都会再转发给VM2 106，两条转发路径没有区别。但是对于像VM1 105和VM3 107这种单归用户设备，流量发往Leaf 102和Leaf 103还是存在区别的。例如，当Spine 101将目的地为VM3 107的流量发送到Leaf 102后，Leaf 102确定VM3 107没有连接到Leaf 102，则Leaf 102通过Peer-link 104将该流量转发给Leaf 103，由Leaf 103将该流量转发给VM3 107，这样的转发路径造成流量不必要的绕行。
图2是本申请M-LAG接入VXLAN的一种报文转发示意图。如图2箭头所示，要发往单归用户设备VM3 107的报文经由Spine 101的隧道接口VTEP1发出，因为Leaf 102和Leaf 103配置了相同的VTEP接口VTEP2，具有相同的VTEP2的IP地址IP2，因此，报文会转发至Leaf 102和Leaf 103中的任意一个，若目的为单归用户设备VM3 107的报文被转发至Leaf 102，则Leaf 102需要通过Peer-link 104将该报文转发到Leaf 103，该报文才可以被正常的转发至单归用户设备VM3 107。在报文转发过程中，Spine 101发送报文时会选择Leaf 102和Leaf 103中的任意一个，因此对于发往单归设备的单播报文而言，将会有一半的概率通过Leaf 102和Leaf 103之间的Peer-link 104，势必加重Peer-link 104的带宽压力，Peer-link需要承担发往单归用户设备的BUM报文和单播报文转发产生的流量，又因为在M-LAG接入VXLAN的网络系统中，BUM报文的比例是很小的，大部分都是发往单归设备的单播报文，因此Peer-link带宽的绝大部分就是被发往单归用户设备的单播报文所占用，因此，大量的发往单归端口的网络侧到用户侧的单播流量，对Peer-link造成巨大的的流量压力。
本申请实施例旨在提供一种单归用户设备的流量优化方案，通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归隧道接口，实现M-LAG接入VXLAN中的单归用户设备的流量优化，使发往单归用户设备的单播流量不会通过M-LAG两台成员设备之间的横向链路Peer-link转发，本申请实施例根据接入设备上用户设备对应的端口的不同类型封装报文，发往不同类型的用户设备的报文具有不同的外层源IP地址，从而能够在不同的VXLAN隧道中精确选择单归隧道接口来转发数据，使数据不经过Peer-link绕行直接转发至该单归用户设备，降低了Peer-link的带宽压力，优化了单归用户设备的转发路径。
图3是本申请实施例的基于M-LAG接入VXLAN的网络系统的一种配置示意图。如图3所示，该网络系统300包括至少一个用户设备(例如，图3中所示的单归用户设备105和107所示的VM1和VM3，双归用户设备106所示的VM2)、骨干设备Spine 101和至少两个接入设备(例如，Leaf 102和Leaf 103)，其中，一个用户设备与至少一个接入设备之间具有通信连接。每个连接有单归用户设备的接入设备中设置有至少一个单归端口(例如，图3中Port3和Port4)，该单归端口Port3仅用于传输需要发送给与接入设备Leaf 102连接的VM1的数据，Port4仅用于传输需要发送给与Leaf 103连接的VM3 107的数据；且每个连接有双归用户设备的接入设备中设置有至少一个双归端口，如图3中的Port1和Port2，该双归端口可以用于传输需要发送给双归用户设备的数据，或用于 传输广播数据，或用于传输需要发送给不与所述接入设备连接的单归用户设备的数据，具体地，例如该双归端口可以用于传输包括广播报文、组播报文以及未知单播报文的BUM报文。
如图所示，图3中的Leaf 102连接有单归用户设备VM1 105，Leaf 103连接有单归用户设备VM3 107，因此为Leaf 102配置VTEP3，在Leaf 103配置VTEP4。VTEP3仅用于传输需要发送给Leaf 102的单归用户设备，例如VM1 105，的数据，VTEP4仅用于传输需要发送给Leaf 103的单归用户设备，例如VM3 107，的数据。
骨干设备Spine 101上配置常规的指向VTEP2的隧道对端(peer)，再额外配置2个M-LAG单归的peer分别指向VTEP3和VTEP4，基于上述配置在Spine 101和接入设备Leaf 102、Leaf 103之间建立隧道，例如隧道VTEP1-VTEP3、VTEP1-VTEP2、VTEP1-VTEP4。
下面，结合图4至图5对本申请实施例的发往单归用户设备的网络侧到用户侧的单播流量的转发方式进行说明。对于用户侧发往网络侧的流量，是负载分担到M-LAG的成员链路，并通过常用链路到达M-LAG的成员设备，然后由成员设备根据本地优先的原则将接收的部分流量转发到网络侧，之后再由网络侧设备转发。因此，在本申请实施例中，重点描述从网络侧到用户侧的报文转发路径。
需要说明的是，以下描述中出现的编号“第一”、“第二”等仅是为了区分不同的描述对象，例如，为了区分不同的用户设备(例如，第一用户设备、第二用户设备)、不同的数据(例如，第一数据、第二数据)、设备标识(例如，第一设备标识、第二设备标识)或隧道接口标识(例如，第一隧道接口标识、第二隧道接口标识)等，不应对本申请实施例的技术方案构成限定。
图4是本申请实施例中传输数据的方法400的示意性流程图。该方法400可以应用于骨干设备中，所述骨干设备连接至少两个接入设备，所述至少两个接入设备组成M-LAG以连接多归用户设备，所述至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述接入设备连接的单归用户设备的数据。
如图4所示，该方法400包括以下内容。
在S410中，骨干设备获得需要发送给第一用户设备的第一数据，根据所述第一映射表，确定所述第一用户设备对应的第一隧道接口标识，其中，所述第一用户设备为单归用户设备。
可选地，所述骨干设备存储有第一映射表，所述第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是所述单归用户设备所连接的接入设备的单归隧道接口的标识。作为另一种实施方式，所述骨干设备获取第一映射表，当所述骨干设备接收到需要发送给单归用户设备的第一数据时，根据所述第一映射表，确定所述单归用户设备对应的单归隧道接口标识。应理解，本发明并不限于此。
以图3为例，第一映射表可以用于指示单归用户设备VM1和VM3对应的隧道接口标识，根据图4所示的方法400，即该第一映射表用于确定单归用户设备VM1对应的隧道接口VTEP3的隧道接口标识，或者单归用户设备VM3对应的隧道接口VTEP4的隧道接口标识。
作为示例而非限定，例如，在本申请实施例中，该第一映射表还可以具体用于指示多个设备标识与多个隧道接口标识之间的映射关系，其中一个设备标识能唯一指示一个隧道接口标识，第一设备标识用于指示第一用户设备。通过每一台用户设备的不同的设备标识来映射不同的隧道接口标识，骨干设备将数据转发至接入设备的单归隧道接口对应的隧道，以将发往单归用户设备的单播报文准确的转发至该单归用户设备所接入的接入设备，实现单归端口的单播流量最优转发的目的。
可选地，例如，单归用户设备VM1的接入端口Port3配置单归VTEP使能属性，VM3的接入端口Port4也配置单归VTEP使能属性，单归用户设备VM1的单归端口使能属性对应于隧道接口VTEP3的隧道接口标识、单归用户设备VM3的单归端口使能属性对应于隧道接口VTEP4的隧道接口标识；或者，单归用户设备VM1的介质访问控制MAC地址对应于隧道接口VTEP3的隧道接口标识、单归用户设备VM3的介质访问控制MAC地址对应于隧道接口VTEP4的隧道接口标识，从而实现骨干设备将数据转发至接入设备的单归隧道接口，以将发往单归用户设备的单播报文准确的转发至该单归用户设备所接入的接入设备，实现单归用户设备的单播流量的最优转发。
在S420中，所述骨干设备根据所述第一隧道接口标识，向第一接入设备发送包括所述第一数据的第一数据包，其中，所述第一接入设备是配置有所述第一隧道接口标识的接入设备。
以图3为例，当骨干设备Spine获得需要发送给VM1(即，第一用户设备)的第一数据时，根据该第一映射表，确定VM1对应的第一隧道接口标识，从而确定报文转发的路径。
可选地，骨干设备Spine根据该第一映射表，将用户设备的设备标识对应的隧道接口标识，确定为第一隧道接口标识，例如将配置有单归端口使能属性的VM1对应的VTEP3隧道接口标识确定为第一隧道接口标识，当骨干设备Spine获得需要发送给VM1的第一数据时，将VTEP1-VTEP3隧道确定为报文转发路径要经过的隧道。
或者，VM1的介质访问控制MAC地址对应于隧道接口VTEP3的隧道接口标识，当报文中包括VM1的MAC地址时，骨干设备Spine将VTEP1-VTEP3的隧道确定为报文转发路径要经过的隧道。
作为示例而非限定，例如，在本申请实施例中，该第一数据可以是骨干设备Spine接收到的包括第一设备标识的第二数据包中的数据。例如，第二数据可以是单播报文、广播报文、组播报文、未知单播报文等报文，第一数据是第二数据包中需要发送到单归用户设备VM1的报文，第一数据包括VM1的设备标识。
作为示例而非限定，该第一设备标识还可以是所述第二数据包中的目的MAC地址。当骨干设备接收到需要发送给与接入设备连接的单归用户设备的数据时，就能根据该目的MAC地址精确选择单归隧道接口VTEP，并根据该单播隧道接口封装该第一数据，并通过该单归隧道接口对应的隧道转发该封装后的第一数据，实现单归用户设备和双归用户设备的转发路径的分离，降低了Peer-link的带宽的压力，优化了单归用户设备的转发路径。
作为示例而非限定，例如，在本申请实施例中，该第一数据可以是单播报文所携带的数据。
在VXLAN中，结合本申请实施例，当Spine获取到原始报文后，会在隧道接口VTEP1进行VXLAN封装，封装得到的VXLAN报文的格式如下表1所示。封装后的VXLAN报文的外部IP报头(Outer IP header)中携带外层源IP地址和外层目的IP地址，其中外层源IP地址为Spine的VTEP的IP地址，外层目的IP地址为目的VM所对应的隧道接口的VTEP IP地址，例如，Spine发往VM1的报文，外层源IP地址为VTEP1的IP地址IP1，外层目的IP地址为单归隧道接口VTEP3的IP地址IP3。Outer IP header的外层目的IP地址可以是单播和组播地址。单播的情况下，外层目的IP地址为VTEP的IP地址，在多播的情况下引入VXLAN的管理层，利用VXLAN标识符(VXLAN Network Identifier，VNI)和IP多播组的映射来确定VTEP。
表1 VXLAN报文格式
Figure PCTCN2019070789-appb-000001
S430，骨干设备Spine根据第一隧道接口标识，通过VXLAN向第一接入设备发送包括第一数据的第一数据包，其中，第一接入设备配置有第一隧道接口标识所指示的单归隧道接口。
在本实施例中，骨干设备Spine根据VTEP3的隧道接口标识，将发往单归用户设备VM1的原始报文进行VXLAN封装，将得到的VXLAN报文通过VTEP1-VTEP3标识的隧道转发给Leaf 102，Leaf 102通过单归隧道接口VTEP3接收该VXLAN报文，解封装该VXLAN报文后，得到发往单归用户设备VM1的原始报文，再通过Leaf 102和VM1的单归端口Port3将该原始报文发送给VM1。
本申请实施例实现了跨设备链路聚合M-LAG接入虚拟可扩展局域网VXLAN的单归接入设备的流量优化，在M-LAG接入VXLAN中，通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归隧道接口，使VXLAN中的主要流量，即发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link进行转发，降低了Peer-link的带宽的压力，优化了单归用户设备的转发路径。
在VXLAN隧道建立后，会在骨干设备和接入设备上下发解封装表项，对应的外层源IP地址是对端的VTEP的IP地址，对应的外层目的IP地址是本端的VTEP的IP地址。当设备(可以是骨干设备或接入设备)收到VXLAN报文后，如果根据该VXLAN外层源IP地址和外层目的IP地址可以命中该设备的解封装表，就表示该设备需要终结这个VXLAN报文，如果不能命中该设备的解封装表，则表明该设备为中转设备，需要继续转发该VXLAN报文到下游设备。按照图3的配置，各设备生成的解封装表如表2所示。
当骨干设备Spine接收到包括广播报文、组播报文、未知单播报文的BUM报文(例如，第二数据包)时，因为VXLAN隧道是一个二层的出端口，因此在VXLAN的广播域中需要加上VXLAN隧道头，让包括广播报文、组播报文和未知单播报文的BUM报文可以在广播域复制到VXLAN隧道中。用于发送BUM报文的表就是VXLAN的头端复制列表。图3 中，Spine 101上指向Leaf 102的peer有2个，M-LAG单归的peer即VTEP1-VTEP3不需要加入头端复制列表，否则会导致Spine 101到Leaf 102的BUM流量出现双份，具体的3个设备上的头端复制列表情况如表2所示。通过VXLAN的解封装列表和头端复制列表的分离，可以防止Spine 101发送BUM报文时，选择单归隧道接口或双归隧道接口的过程中在不同的VXLAN隧道上向同一个接入设备复制多次BUM报文的问题。
表2各设备的解封装表项、头端复制列表对比
Figure PCTCN2019070789-appb-000002
在现有技术中，Spine通过同一个隧道接口发送目的地为单归用户设备和双归用户设备的报文。如果Spine将目的地为单归用户设备的报文发给了与该单归用户设备的接入设备不同的另一接入设备，该另一接入设备就需要通过接入设备之间的横向链路Peer-link将该报文转发给该单归用户设备的接入设备，这会导致横向链路Peer-link的巨大流量压力。本申请实施例通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归隧道接口，使得单播流量通过单归隧道接口转发到该单归用户设备所连接的接入设备来实现单归用户设备的单播流量的最优转发，使发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link转发，降低了Peer-link的带宽的压力，优化了单归用户设备的转发路径。接入设备原有的隧道接口(即双归隧道接口)，用来传输需要发送给双归用户设备的数据、广播数据和需要发送给不与该接入设备连接的单归用户设备的数据中的至少一种，从而实现路径的分离，降低了带宽的压力。
图5是本申请实施例中传输数据的方法500的又一示意性框图。该方法500可以应用于第一接入设备中，所述第一接入设备与第二接入设备组成M-LAG以连接多归用户设备，所述第一接入设备还用于连接单归用户设备，所述第一接入设备中设置有单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述第一接入设备连接的单归用户设备的数据，但本申请实施例不限于此。
如图5所示，该方法500包括以下内容。
在S510中，所述第一接入设备通过第一单归隧道接口接收包括第一数据的第一数据包，其中，所述第一数据的目的为第一用户设备，所述第一单归隧道接口是第一隧道接口标识指示的单归隧道接口，所述第一隧道接口标识是所述第一用户设备对应的隧道接口标识，所述第一用户设备为与所述第一接入设备连接的单归用户设备。
以图3为例，S510具体可以是骨干设备Spine 101将发往单归用户设备VM1的报文 (第一数据)封装在VXLAN报文(第一数据包)中发送给Leaf 102，将该VXLAN报文通过VTEP1-VTEP3标识的隧道转发给Leaf 102，Leaf 102通过VTEP3接收该VXLAN报文。
Leaf 102需要通过隧道接口VTEP3来接收要发往与Leaf 102连接的用户设备VM1的报文。则Leaf 102首先要使Spine101学习到VTEP3。Leaf 102在向Spine 101转发VM1发送的报文的时候，根据Leaf 102连接VM1的端口(Port 3)是否配置单归VTEP使能属性选择外层源IP地址，当Port 3配置了单归VTEP使能的情况下确定的外层源IP地址是VTEP3的IP地址(记为VTEP3)，否则确定的外层源IP地址是VTEP2。Leaf 103上的处理类似，不再详细描述。Spine101接收到Leaf 102发送的VM1的报文时，学习到VM1对应的隧道出接口为VTEP3，Spine101接收到Leaf 102或Leaf 103发送的VM2的报文时，学习到VM2对应的隧道出接口为VTEP2，Spine101接收到Leaf 103发送的VM3的报文时，学习到VM3对应的隧道出接口为VTEP4。通过上述过程，Spine 101上学习到的转发表如表3所示。
表3 Spine上的转发表项
VM 转发表项出端口
VM1 VXLAN TUNNEL(VTEP3-VTEP1)
VM2 VXLAN TUNNEL(VTEP2-VTEP1)
VM3 VXLAN TUNNEL(VTEP4-VTEP1)
经过上述学习后，骨干设备发往接入设备的单播报文，例如从Spine 101发往Leaf102的流量会存在2种封装的类型，外层目的IP地址分别是VTEP2和VTEP3。根据用户设备类型的不同封装不同的VXLAN头中的源IP地址，可以在不分离转发表项的情况下，达到单归用户设备和双归用户设备的转发路径分离的目的。
骨干设备发往接入设备的广播报文、组播报文或未知单播报文，例如从Spine 101发往Leaf 102和Leaf 103的BUM报文在Spine 101上的处理有所区别。参见表2的头端复制列表，Spine 101只会将该BUM报文复制一份发往VTEP2，不向单归的VTEP3、VTEP4进行复制。
在S520中，所述第一接入设备处理所述第一数据包得到所述第一数据。前面描述了在VXLAN中，报文在隧道接口进行封装，第一接入设备接收到的第一数据包包括VXLAN的隧道头，因此第一接入设备需要对第一数据包进行解封装，得到需要发送给第一用户设备的第一数据。
在S530中，所述第一接入设备向所述第一用户设备发送所述第一数据。
具体地，以图3为例，可以是Leaf1向VM1发送所述第一数据。
在现有技术中，发往单归用户设备和双归用户设备的不同类型的数据都是通过一个端口进行转发，对单归用户设备而言，如果发往不是自己接入的接入设备，就需要通过接入设备之间的横向链路Peer-link进行转发，所以才会导致横向链路Peer-link的巨大流量压力，本申请实施例将发往单归用户设备的单播流量和其他流量路径进行区分，通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用 户设备数据的单归端口，使得单播流量通过单归隧道接口转发到该单归用户设备所连接的接入设备来实现单归用户设备的流量优化，使发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link转发，降低了Peer-link的带宽的压力，优化了单播流量的转发路径，实现单归用户设备的单播流量的最优转发。对于接入设备原有的端口(即双归端口)，用来传输需要发送给双归用户设备的数据，和/或该双归端口用于传输广播数据，和/或该双归端口用于传输需要发送给不与该接入设备连接的单归用户设备的数据，从而实现路径的分离，降低了带宽的压力。
本申请实施例提供的方案实现了M-LAG接入VXLAN的单归接入设备的流量优化，通过在单归用户设备所连接的接入设备设置至少一个仅用于传输需要发送给该单归用户设备数据的单归端口，实现单归用户设备的流量优化，使VXLAN中的主要流量，发往单归用户设备的单播流量不会通过M-LAG多台成员设备之间即接入设备之间的横向链路Peer-link进行转发，而是根据端口的类型封装报文，使报文具有不同的源IP，从而能够实现在不同的VXLAN隧道中精确选择单归隧道接口来转发数据，使数据不经过Peer-link绕行直接转发至该单归用户设备，降低了Peer-link的带宽的压力，优化了单归端口的转发路径，实现单播流量最优转发的目的。
图6是本申请实施例中骨干设备的示意性框图。具体地，该骨干设备600配置在包括至少两个接入设备的网络系统中，该至少两个接入设备之间具有通信连接，用于连接多归用户设备，且该至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个单归用户设备仅连接一个接入设备，每个连接有单归用户设备的接入设备中设置有至少一个仅用于传输需要发送给与该接入设备连接的单归用户设备的数据的单归隧道接口。
在本申请实施例中，该骨干设备600可以包括：存储单元610、处理单元620和通信单元630。
其中，存储单元610用于存储第一映射表，该第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是所述单归用户设备所连接的接入设备的单归隧道接口的标。
处理单元620用于在通信单元630获得需要发送给该单归用户设备的第一数据时，根据第一映射表，确定该单归用户设备对应的隧道接口标识。
通信单元630用于根据处理单元620确定的隧道接口标识，通过该隧道接口标识所指示的单归隧道接口，向该单归用户设备的接入设备发送包括有所述第一数据的第一数据包。
应理解，该骨干设备600可以对应(例如，可以配置于或本身即为)上述方法400中描述的骨干设备Spine，并且，该设备600中各个模块或单元分别用于执行上述方法400中骨干设备Spine所执行的各动作或处理过程。
还应理解，图6示出的传输数据的装置600仅为示例，本申请实施例的传输数据的装置还可包括其他模块或单元，或者包括与图6中的各个模块的功能相似的模块，或者并非要包括图6中的所有模块。
图7是本申请实施例中接入设备的示意性框图。该接入设备配置在包括骨干设备和至少两个该接入设备的网络系统中，其中，至少两个该接入设备之间具有通信连接，用 于连接多归用户设备，该至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个单归用户设备仅连接一个该接入设备，每个连接有单归用户设备的接入设备中设置有至少一个仅用于传输需要发送给与该接入设备连接的单归用户设备的数据的单归隧道接口。
在本申请实施例中，该接入设备700可以包括：第一通信单元710、处理单元720和第二通信单元730。
其中，第一通信单元710具体用于通过单归隧道接口接收包括第一数据的第一数据包，其中，所述第一数据的目的为单归用户设备，该单归隧道接口是隧道接口标识指示的单归隧道接口，所述隧道接口标识是所述单归用户设备对应的隧道接口标识。
处理单元720用于处理所述第一数据包，得到所述第一数据。前面描述了在VXLAN中，报文在隧道接口进行封装，第一通信单元710接收到的第一数据包包括VXLAN的隧道头，因此处理单元720需要对第一数据包进行解封装，得到需要发送给第一用户设备的第一数据。
第二通信单元730具体用于向所述单归用户设备发送第一数据。
应理解，该接入设备700可以对应(例如，可以配置于或本身即为)上述方法500中描述的接入设备，并且，该设备700中各个模块或单元分别用于执行上述方法500中接入设备所执行的各动作或处理过程，这里，为了避免赘述，省略其详细说明。
还应理解，图7示出的传输数据的装置700仅为示例，本申请实施例的传输数据的装置还可包括其他模块或单元，或者包括与图7中的各个模块的功能相似的模块，或者并非要包括图7中的所有模块。
图8是本申请实施例提供的一种传输数据的设备800的示意性框图。该设备可以为上述方法中的骨干设备或接入设备，该设备800包括：处理器810、存储器820、通信接口830和总线840。其中，处理器810、存储器820、通信接口830通过总线840通信，也可以通过无线传输等其他手段通信。该存储器820用于存储指令，该处理器810用于执行该存储器820存储的指令，该存储器820存储程序代码，且处理器810可以调用存储器820中存储的程序代码，以控制通信接口830收发信息或信号，使得设备800执行上述方法400中骨干设备中各处理单元的功能、所执行的动作或处理过程，或执行上述方法500中接入设备中各处理单元的功能、所执行的动作或处理过程。这里，为了避免赘述，省略其详细说明。
若该设备800为骨干设备，则该存储器820还存储有第一映射表，所述第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是所述单归用户设备所连接的接入设备的单归隧道接口的标识。处理器810可用于在获得需要发送给第一用户设备的第一数据时，根据所述第一映射表，确定所述第一用户设备对应的第一隧道接口标识，所述第一用户设备为单归用户设备；该处理器810还可用控制该通信接口830根据所述第一隧道接口标识，向第一接入设备发送包括所述第一数据的第一数据包，其中，所述第一接入设备是配置有所述第一隧道接口标识所指示的单归端口的接入设备。
若该设备800为接入设备，则该通信接口830可用于通过单归隧道接口(例如可对应于网络设备900中的以太网接口914)接收包括第一数据的第一数据包，由处理器810 处理处理所述第一数据包，得到所述第一数据，通过通信接口830向单归用户设备发送所述第一数据，其中，所述第一数据的目的为单归用户设备，所述单归隧道接口是隧道接口标识指示的单归隧道接口，所述隧道接口标识是所述单归用户设备对应的隧道接口标识。
图9是本申请实施例提供的另一种传输数据的设备900的示意性框图。如图9所示，该设备900(例如，骨干设备或接入设备)通常包括接口板910和主控板920，接口板910和主控板920之间可通过内部总线930连接。接口板910包括中央处理器(central processing unit，CPU)911、存储器912、以太交换芯片(LAN switch，LSW)913、和以太网接口914等，主控板920包括CPU 921、存储器922等。
其中，主控板系统内存和接口板的板卡内存共同映射在该网络设备的内存空间中，CPU是网络设备主控板和接口板的控制单元，存储器例如可以包括随机存储器(random access memory，RAM)和/或闪存(Flash)等，运行的程序和静态配置的参数可以保存在闪存中，程序运行时执行的代码和数据可以放置在与CPU相连的RAM中。可选地，LSW 913可配有外置RAM，存放转发的报文内容，可以解决LSW芯片内部缓存不足的某些场景，LSW 913下挂的物理层PHY完成设备之间的光口或电口的以太网接口对接。
在报文转发过程中，由网络设备的接口板910的以太交换芯片LSW 913通过以太网接口914从该网络设备的外部接收到以太数据报文，LSW 913接收CPU的配置指令、转发表项等，指导LSW 913对报文的转发行为。
在本申请实施例中，该CPU 921和CPU 911可对应于上述设备800中的处理器810，该LSW 913和以太网接口914可对应于上述设备800中的通信接口830。各模块或单元的具体功能在上文中已经详细说明，为了简洁，这里不再赘述。
本申请实施例通过主控板和接口板中的CPU的协同合作，从而控制LSW的转发行为，实现了M-LAG接入VXLAN的单归用户设备的流量优化，可以使VXLAN中的主要流量，即发往单归用户设备的单播流量不会通过Peer-link转发，降低Peer-link的带宽的压力。
应注意，上述方法实施例可以应用于处理器中，或者由处理器实现。处理器可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法实施例的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。
可以理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器 (read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。应注意，本文描述的系统和方法的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
应理解，在本发明实施例的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本发明实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，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)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (18)

  1. 一种传输数据的方法，其特征在于，应用于骨干设备中，所述骨干设备连接至少两个接入设备，所述至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，所述至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述接入设备连接的单归用户设备的数据，所述骨干设备存储有第一映射表，所述第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是所述单归用户设备所连接的接入设备的单归隧道接口的标识，所述方法包括：
    当所述骨干设备获得需要发送给第一用户设备的第一数据时，根据所述第一映射表，确定所述第一用户设备对应的第一隧道接口标识，其中，所述第一用户设备为单归用户设备；
    所述骨干设备根据所述第一隧道接口标识，向第一接入设备发送包括所述第一数据的第一数据包，其中，所述第一接入设备是配置有所述第一隧道接口标识的接入设备。
  2. 根据权利要求1所述的方法，其特征在于，所述骨干设备和所述至少两个接入设备中的每个接入设备支持虚拟可扩展局域网络VXLAN功能，
    所述骨干设备根据所述第一隧道接口标识，向第一接入设备发送包括所述第一数据的第一数据包，包括：
    所述骨干设备生成所述第一数据包，所述第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括所述第一隧道接口标识；
    所述骨干设备通过所述第一隧道接口标识对应的VXLAN隧道向所述第一接入设备发送所述第一数据包。
  3. 根据权利要求1或2所述的方法，其特征在于，所述第一映射表具体用于指示所述多个单归用户设备中每个单归用户设备的标识与所述单归用户设备的隧道接口标识之间的映射关系；
    所述根据所述第一映射表，确定所述第一用户设备对应的第一隧道接口标识，包括：
    所述骨干设备根据所述第一映射表，将第一设备标识对应的隧道接口标识，确定为所述第一隧道接口标识，所述第一设备标识用于指示所述第一用户设备。
  4. 根据权利要求3所述的方法，其特征在于，所述第一数据是所述骨干设备接收到的第二数据包中携带的数据，并且，所述第二数据包中包括所述第一设备标识。
  5. 根据权利要求1至4中任一项所述的方法，其特征在于，所述骨干设备通过单播方式获得所述第一数据。
  6. 一种传输数据的方法，其特征在于，应用于第一接入设备中，所述第一接入设备与第二接入设备组成跨设备链路聚合组以连接多归用户设备，所述第一接入设备还用于连接单归用户设备，所述第一接入设备中设置有单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述第一接入设备连接的单归用户设备的数据，所述方法包括：
    所述第一接入设备通过第一单归隧道接口接收包括第一数据的第一数据包，其中，所述第一数据的目的为第一用户设备，所述第一单归隧道接口是第一隧道接口标识指示 的单归隧道接口，所述第一隧道接口标识是所述第一用户设备对应的隧道接口标识，所述第一用户设备为与所述第一接入设备连接的单归用户设备；
    所述第一接入设备处理所述第一数据包得到所述第一数据；
    所述第一接入设备向所述第一用户设备发送所述第一数据。
  7. 根据权利要求6所述的方法，其特征在于，所述第一接入设备和第二接入设备支持虚拟可扩展局域网络VXLAN功能，
    所述第一接入设备通过第一单归隧道接口接收包括第一数据的第一数据包，包括：
    所述第一接入设备通过所述第一单归隧道接口对应的VXLAN隧道接收所述第一数据包，所述第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括所述第一隧道接口标识。
  8. 根据权利要求6或7所述的方法，其特征在于，所述第一接入设备和第二接入设备中设置有共同的双归隧道接口，
    所述双归隧道接口用于传输需要发送给双归用户设备的数据，和/或
    所述双归隧道接口用于传输广播数据，和/或
    所述双归隧道接口用于传输需要发送给不与所述接入设备连接的单归用户设备的数据。
  9. 一种骨干设备，其特征在于，所述骨干设备用于连接至少两个接入设备，所述至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，所述至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述接入设备连接的单归用户设备的数据，所述骨干设备包括：
    存储单元，用于存储第一映射表，所述第一映射表用于指示多个单归用户设备中的每个单归用户设备对应的隧道接口标识，其中，一个单归用户设备对应的隧道接口标识是所述单归用户设备所连接的接入设备的单归隧道接口的标识；
    处理单元，用于在获得需要发送给第一用户设备的第一数据时，根据所述第一映射表，确定所述第一用户设备对应的第一隧道接口标识，所述第一用户设备为单归用户设备；
    通信单元，用于根据所述第一隧道接口标识，向第一接入设备发送包括所述第一数据的第一数据包，其中，所述第一接入设备是配置有所述第一隧道接口标识所指示的单归隧道接口的接入设备。
  10. 根据权利要求9所述的骨干设备，其特征在于，所述骨干设备和所述至少两个接入设备中的每个接入设备支持虚拟可扩展局域网络VXLAN功能，
    所述处理单元具体用于生成所述第一数据包，所述第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括所述第一隧道接口标识；
    所述通信单元具体用于通过所述第一隧道接口标识对应的VXLAN隧道向所述第一接入设备发送所述第一数据包。
  11. 根据权利要求9或10所述的骨干设备，其特征在于，所述第一映射表具体用于指示多个单归用户设备中每个单归用户设备的标识与所述单归用户设备的隧道接口标识之间的映射关系，
    所述处理单元具体用于根据所述第一映射表，将第一设备标识对应的隧道接口标识，确定为所述第一隧道接口标识，所述第一设备标识用于指示所述第一用户设备。
  12. 根据权利要求11所述的骨干设备，其特征在于，所述第一数据是所述骨干设备接收到的第二数据包中携带的数据，并且，所述第二数据包中包括所述第一设备标识。
  13. 根据权利要求9至12中任一项所述的骨干设备，其特征在于，所述骨干设备通过单播方式获得所述第一数据。
  14. 一种接入设备，其特征在于，所述接入设备为第一接入设备，所述第一接入设备与第二接入设备组成跨设备链路聚合组以连接多归用户设备，所述第一接入设备还用于连接单归用户设备，所述第一接入设备中设置有单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述第一接入设备连接的单归用户设备的数据，所述第一接入设备包括：
    第一通信单元，用于通过第一单归隧道接口接收包括第一数据的第一数据包，其中，所述第一数据的目的为第一用户设备，所述第一单归隧道接口是第一隧道接口标识指示的单归隧道接口，所述第一隧道接口标识是所述第一用户设备对应的隧道接口标识，所述第一用户设备为与所述第一接入设备连接的单归用户设备；
    处理单元，用于处理所述第一数据包，得到所述第一数据；
    第二通信单元，用于向所述第一用户设备发送所述第一数据。
  15. 根据权利要求14所述的接入设备，其特征在于，所述第一接入设备和第二接入设备支持虚拟可扩展局域网络VXLAN功能，
    所述通信单元具体用于通过所述第一单归隧道接口对应的VXLAN隧道接收所述第一数据包，所述第一数据包包括VXLAN隧道头，所述VXLAN隧道头的外层源地址字段包括所述第一隧道接口标识。
  16. 根据权利要求14或15所述的接入设备，其特征在于，所述第一接入设备和第二接入设备中设置有共同的双归隧道接口，
    所述双归隧道接口用于传输需要发送给双归用户设备的数据，和/或
    所述双归隧道接口用于传输广播数据，和/或
    所述双归隧道接口用于传输需要发送给不与所述装置连接的单归用户设备的数据。
  17. 一种网络系统，包括：
    如权利要求9至13中任一项所述的骨干设备；和
    至少两个如权利要求14至16中任一项所述的接入设备；
    所述骨干设备连接所述至少两个接入设备，所述至少两个接入设备组成跨设备链路聚合组以连接多归用户设备，所述至少两个接入设备中的至少一个接入设备还用于连接单归用户设备，每个连接有单归用户设备的接入设备中设置有至少一个单归隧道接口，所述单归隧道接口仅用于传输需要发送给与所述接入设备连接的单归用户设备的数据。
  18. 一种计算机可读存储介质，其特征在于，用于存储计算机指令，当所述计算机指令在计算机上运行时，使得所述计算机执行如权利要求1-9中任一项所述传输数据的方法。
PCT/CN2019/070789 2018-01-11 2019-01-08 传输数据的方法、设备和网络系统 WO2019137355A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP23158941.7A EP4221107A3 (en) 2018-01-11 2019-01-08 Data transmission method, device, and network system
EP19738772.3A EP3731472B1 (en) 2018-01-11 2019-01-08 Method and device for transmitting data, and network system
US16/924,734 US11539551B2 (en) 2018-01-11 2020-07-09 Data transmission method, device, and network system
US17/980,631 US20230073291A1 (en) 2018-01-11 2022-11-04 Data transmission method, device, and network system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810026227.1A CN110034993B (zh) 2018-01-11 2018-01-11 传输数据的方法、设备和网络系统
CN201810026227.1 2018-01-11

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/924,734 Continuation US11539551B2 (en) 2018-01-11 2020-07-09 Data transmission method, device, and network system

Publications (1)

Publication Number Publication Date
WO2019137355A1 true WO2019137355A1 (zh) 2019-07-18

Family

ID=67218903

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/070789 WO2019137355A1 (zh) 2018-01-11 2019-01-08 传输数据的方法、设备和网络系统

Country Status (4)

Country Link
US (2) US11539551B2 (zh)
EP (2) EP3731472B1 (zh)
CN (1) CN110034993B (zh)
WO (1) WO2019137355A1 (zh)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110034993B (zh) * 2018-01-11 2020-12-25 华为技术有限公司 传输数据的方法、设备和网络系统
CN110543340A (zh) * 2019-08-09 2019-12-06 北京信安世纪科技股份有限公司 一种确定接口库对应密码设备的方法、装置、设备及介质
CN112769587B (zh) * 2019-11-05 2024-05-14 中兴通讯股份有限公司 双归设备接入流量的转发方法、设备及存储介质
CN113381929B (zh) * 2020-03-10 2023-03-24 华为技术有限公司 一种路由处理方法、网关设备及计算机存储介质
CN111585863B (zh) * 2020-06-11 2022-03-01 国家计算机网络与信息安全管理中心 虚拟可扩展局域网报文处理设备及其数据处理方法
CN111865779B (zh) * 2020-07-13 2022-08-02 中国联合网络通信集团有限公司 一种路由同步方法及跨设备链路聚合组
CN112291160B (zh) * 2020-09-23 2022-03-01 浪潮思科网络科技有限公司 一种bum报文抑制方法、设备及介质
CN113259435B (zh) * 2021-05-13 2022-07-12 上海巨印科技有限公司 辐射测量仪数据传输方法及系统
CN115051890A (zh) * 2022-05-20 2022-09-13 中国电信股份有限公司 一种报文处理方法、系统、装置、电子设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103957160A (zh) * 2014-05-12 2014-07-30 华为技术有限公司 一种发送报文的方法及设备
CN106936939A (zh) * 2015-12-31 2017-07-07 华为技术有限公司 一种报文处理方法、相关装置及nvo3网络系统
CN106998286A (zh) * 2017-05-05 2017-08-01 杭州迪普科技股份有限公司 一种vxlan报文转发方法及装置
CN107135118A (zh) * 2016-02-29 2017-09-05 华为技术有限公司 一种单播通信方法、网关以及vxlan接入设备

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9350608B2 (en) * 2014-01-10 2016-05-24 Arista Networks, Inc. Method and system for using virtual tunnel end-point registration and virtual network identifiers to manage virtual extensible local area network access
CN105634942B (zh) * 2014-10-31 2020-01-03 华为技术有限公司 转发报文的方法和交换机
CN105743780B (zh) * 2014-12-09 2019-05-28 华为技术有限公司 报文传输方法及装置
US10171306B2 (en) * 2015-02-26 2019-01-01 Cisco Technology, Inc. Automatic discovery and provisioning of multi-chassis etherchannel peers
US10200278B2 (en) * 2016-03-02 2019-02-05 Arista Networks, Inc. Network management system control service for VXLAN on an MLAG domain
US10142129B1 (en) * 2016-09-27 2018-11-27 Juniper Networks, Inc. Bum packet filtering in multi-homed EVPN overlay networks
US10237090B2 (en) * 2016-10-28 2019-03-19 Avago Technologies International Sales Pte. Limited Rule-based network identifier mapping
CN107547366B (zh) * 2017-05-15 2020-10-27 新华三技术有限公司 一种报文转发方法和装置
US10164873B1 (en) * 2017-06-01 2018-12-25 Ciena Corporation All-or-none switchover to address split-brain problems in multi-chassis link aggregation groups
US10659254B2 (en) * 2017-07-10 2020-05-19 Fungible, Inc. Access node integrated circuit for data centers which includes a networking unit, a plurality of host units, processing clusters, a data network fabric, and a control network fabric
CN107547402B (zh) * 2017-07-19 2020-04-03 新华三技术有限公司 一种转发表生成方法和装置
US10425325B2 (en) * 2017-10-30 2019-09-24 Dell Products Lp Optimizing traffic paths to orphaned hosts in VXLAN networks using virtual link trunking-based multi-homing
CN110034993B (zh) * 2018-01-11 2020-12-25 华为技术有限公司 传输数据的方法、设备和网络系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103957160A (zh) * 2014-05-12 2014-07-30 华为技术有限公司 一种发送报文的方法及设备
CN106936939A (zh) * 2015-12-31 2017-07-07 华为技术有限公司 一种报文处理方法、相关装置及nvo3网络系统
CN107135118A (zh) * 2016-02-29 2017-09-05 华为技术有限公司 一种单播通信方法、网关以及vxlan接入设备
CN106998286A (zh) * 2017-05-05 2017-08-01 杭州迪普科技股份有限公司 一种vxlan报文转发方法及装置

Also Published As

Publication number Publication date
EP3731472B1 (en) 2023-03-29
CN110034993A (zh) 2019-07-19
EP3731472A1 (en) 2020-10-28
CN110034993B (zh) 2020-12-25
EP4221107A2 (en) 2023-08-02
EP4221107A3 (en) 2023-09-13
EP3731472A4 (en) 2021-01-27
US20230073291A1 (en) 2023-03-09
US20200374155A1 (en) 2020-11-26
US11539551B2 (en) 2022-12-27

Similar Documents

Publication Publication Date Title
WO2019137355A1 (zh) 传输数据的方法、设备和网络系统
US11765000B2 (en) Method and system for virtual and physical network integration
EP3379806B1 (en) Packet processing method, related apparatus and nvo3 network system
US9698995B2 (en) Systems and methods for providing multicast routing in an overlay network
US9912612B2 (en) Extended ethernet fabric switches
CN103200069B (zh) 一种报文处理的方法和设备
US9374323B2 (en) Communication between endpoints in different VXLAN networks
US8774038B2 (en) Multicast support for dual stack-lite and internet protocol version six rapid deployment on internet protocol version four infrastructures
US10193707B2 (en) Packet transmission method and apparatus
CN103841023B (zh) 数据转发的方法和设备
JP6722816B2 (ja) パケット転送
WO2015123987A1 (zh) 报文转发方法和装置
WO2013029440A1 (en) Method and apparatus for implementing layer-2 interconnection of data centers
US20170078115A1 (en) Multi-homed access
CN105187311B (zh) 一种报文转发方法及装置
CN107659484B (zh) 从vlan网络接入vxlan网络的方法、装置及系统
CN107623636B (zh) 一种用户隔离方法和交换机
WO2022117018A1 (zh) 报文传输的方法和装置
JP2019526208A (ja) 機器検知
WO2017036384A1 (zh) 运营商边缘设备及数据转发方法
WO2018233632A1 (zh) 视频点播流的转发
JP2017098738A (ja) 制御装置、通信システム、制御方法およびプログラム
CN110324247A (zh) 三层组播网络中的组播转发方法、设备及存储介质
CN115695087A (zh) 跨域局域网的建立方法、装置、设备及存储介质

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: 19738772

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: 2019738772

Country of ref document: EP

Effective date: 20200721