WO2016019822A1 - 一种分流流量的方法和装置 - Google Patents

一种分流流量的方法和装置 Download PDF

Info

Publication number
WO2016019822A1
WO2016019822A1 PCT/CN2015/085441 CN2015085441W WO2016019822A1 WO 2016019822 A1 WO2016019822 A1 WO 2016019822A1 CN 2015085441 W CN2015085441 W CN 2015085441W WO 2016019822 A1 WO2016019822 A1 WO 2016019822A1
Authority
WO
WIPO (PCT)
Prior art keywords
network node
bandwidth
forwarding
forwarding tunnel
packet loss
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2015/085441
Other languages
English (en)
French (fr)
Inventor
郭尧亮
唐亮
蒋维廉
柯海荣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP15829712.7A priority Critical patent/EP3166267B1/en
Publication of WO2016019822A1 publication Critical patent/WO2016019822A1/zh
Priority to US15/423,922 priority patent/US10251091B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/08—Load balancing or load distribution
    • H04W28/082—Load balancing or load distribution among bearers or channels
    • 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
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/0231—Traffic management, e.g. flow control or congestion control based on communication conditions
    • H04W28/0236—Traffic management, e.g. flow control or congestion control based on communication conditions radio quality, e.g. interference, losses or delay
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • H04W28/18—Negotiating wireless communication parameters
    • H04W28/20—Negotiating bandwidth
    • 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
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • 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
    • H04L47/26—Flow control; Congestion control using explicit feedback to the source, e.g. choke packets
    • H04L47/263—Rate modification at the source after receiving feedback
    • 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
    • H04L47/29—Flow control; Congestion control using a combination of thresholds

Definitions

  • the present invention relates to the field of communications, and in particular, to a method and apparatus for splitting traffic.
  • Hybrid Access is a relatively common way for user terminals to connect to a network.
  • the so-called hybrid access method means that user terminals are connected to the network through multiple different types of data transmission links.
  • a typical application scenario in the hybrid access mode is that the user terminal is connected to the network through a Digital Subscriber Line (DSL) link and a Long Term Evolution (LTE) link.
  • DSL Digital Subscriber Line
  • LTE Long Term Evolution
  • FIG. 1 is a schematic diagram of a network topology of a hybrid access network.
  • a PC Personal Computer
  • HG Home Gateway
  • HAAP Aggregation Point
  • the user terminal first uses the DSL link to exchange data with the network. When the traffic between the user terminal and the network exceeds 50M that the DSL link can carry, Overflow occurs, and traffic beyond the bandwidth of the DSL link will be offloaded to the LTE link for forwarding.
  • the bandwidth on the DSL link between HG1 and HG2 and the Digital Subscriber Line Access Multiplexer (DSLAM) is guaranteed, but the DSLAM passes through the router to the HAAP.
  • the IP link is provided, and the bandwidth provided to each user terminal cannot be guaranteed.
  • the IP is In the process of forwarding traffic on the link, serious packet loss still occurs, which affects network quality.
  • the present invention provides a method and apparatus for splitting traffic to reduce packet loss.
  • an embodiment of the present invention provides a method for offloading traffic, where a first data link and a second data link are connected between a first network node and a second network node, where the first data link is Establishing a first forwarding tunnel from the first network node to the second network node, where the first network node and the second network node are respectively tunnel endpoints of the first forwarding tunnel, including:
  • the first network node calculates a first packet loss rate of the first forwarding tunnel in a preset period
  • the first network node When the first packet loss rate is greater than the first predetermined threshold, the first network node reduces the rated bandwidth of the first forwarding tunnel by a first predetermined bandwidth to obtain a first available bandwidth;
  • the first network node When the forwarding traffic of the first forwarding tunnel exceeds the first available bandwidth, the first network node offloads the excess forwarding traffic to the second data link for forwarding.
  • the method further includes:
  • the first network node When the first packet loss rate is less than a second preset threshold, the first network node increases the first available bandwidth by a second predetermined bandwidth to obtain a second available bandwidth, where the second predetermined bandwidth is less than or equal to Describe a first predetermined bandwidth, the second available bandwidth being less than or equal to a nominal bandwidth of the first data link from the first network node to a second network node.
  • the first network node calculates a first packet loss of the first forwarding tunnel in a preset period Rate, including:
  • the first network node sends a first count message to the second network node, and the first count message carries the first network node in the preset period. Sending, by the first forwarding tunnel, a total number of data packets to the second network node;
  • a second forwarding tunnel from the second network node to the first network node is further established in the first data link, where the first network node and The second network node is divided into
  • the tunnel endpoint of the second forwarding tunnel includes:
  • the first network node calculates a second packet loss rate of the second forwarding tunnel in the preset period
  • the first network node When the second packet loss rate is greater than a third predetermined threshold, the first network node sends a bandwidth advertisement message to the second network node, where the bandwidth advertisement message is used to indicate that the second network node
  • the rated bandwidth of the second forwarding tunnel is reduced by a third predetermined bandwidth to obtain a third available bandwidth.
  • the method further includes:
  • the second network node When the forwarding traffic of the second forwarding tunnel exceeds the third available bandwidth, the second network node offloads the excess forwarding traffic to the second data link for forwarding.
  • the first data link is a digital subscriber line DSL link and a network protocol IP link;
  • the second data link is a Long Term Evolution (LTE) link.
  • LTE Long Term Evolution
  • the second network node is a hybrid access convergence point HAAP;
  • the second network node is an HG.
  • an embodiment of the present invention provides a device for offloading traffic, where a first data link and a second data link are connected between a first network node and a second network node, where the first data link is Establishing a first forwarding tunnel from the first network node to the second network node, where the first network node and the second network node are respectively tunnel endpoints of the first forwarding tunnel, including:
  • a first calculating unit configured to calculate a first packet loss rate of the first forwarding tunnel in a preset period
  • a bandwidth reduction unit configured to reduce a nominal bandwidth of the first forwarding tunnel by a first predetermined bandwidth when the execution result of the first computing unit is that the first packet loss rate is greater than a first predetermined threshold, to obtain a first Available bandwidth
  • the first offloading unit is configured to: when the forwarding traffic of the first forwarding tunnel exceeds the first available bandwidth, offload the excess forwarding traffic to the second data link for forwarding.
  • the device further includes:
  • a bandwidth increasing unit configured to: when the first computing unit performs an execution result that the first packet loss rate is less than a second preset threshold, increase the first available bandwidth by a second predetermined bandwidth to obtain a second available bandwidth.
  • the second predetermined bandwidth is less than or equal to the first predetermined bandwidth
  • the second available bandwidth is less than or equal to the first data link. The nominal bandwidth of the first network node to the second network node.
  • the first calculating unit includes:
  • a first counting packet sending subunit configured to send a first counting packet to the second network node every the preset period, where the first counting packet carries the first network node Transmitting, by the first forwarding tunnel, the total number of data packets to the second network node in a preset period;
  • a second counting message receiving subunit configured to receive a second counting message returned by the second network node, where the second counting message carries the first network carried in the first counting message Sending, by the first forwarding tunnel, the total number of data packets to the second network node in the preset period, and receiving, by the second network node, the first forwarding tunnel in the preset period Describe the total number of data packets sent by the first network node;
  • a first packet loss rate calculation subunit configured to receive, according to the total number of data packets sent by the first network node to the second network node, and the second network node, according to the second count message And a total number of data packets sent by the first network node, and calculating a first packet loss rate of the first forwarding tunnel corresponding to the preset period.
  • a second forwarding tunnel from the second network node to the first network node is further established in the first data link, where the first network node and The second network node is a tunnel endpoint of the second forwarding tunnel, and the device includes:
  • a second calculating unit configured to calculate a second packet loss rate of the second forwarding tunnel in the preset period
  • a sending unit configured to send a bandwidth advertisement message to the second network node, where the second packet loss rate is greater than a third predetermined threshold, and the bandwidth advertisement packet is used by the second calculation unit
  • the second network node is instructed to reduce the rated bandwidth of the second forwarding tunnel by a third predetermined bandwidth to obtain a third available bandwidth.
  • the apparatus further includes:
  • the second offloading unit is configured to: when the forwarding traffic of the second forwarding tunnel exceeds the third available bandwidth, offload the excess forwarding traffic to the second data link for forwarding.
  • the first data link is a digital subscriber line DSL link and a network protocol IP link;
  • the second data link is a Long Term Evolution (LTE) link.
  • LTE Long Term Evolution
  • the second network node is a hybrid access convergence point HAAP;
  • the second network node is an HG.
  • the first network node reduces the available bandwidth of the first forwarding tunnel to the first available bandwidth when the packet loss ratio of the first forwarding tunnel is large, so that when the first When the traffic forwarded by the forwarding tunnel becomes larger, once the first available bandwidth is exceeded, the excess forwarding traffic is started to be offloaded to the second data link for forwarding, because the first available bandwidth is smaller than the original first shown.
  • the available bandwidth of the forwarding tunnel the forwarding traffic will exceed the first available bandwidth earlier, and the shunting time of the traffic of the first network node is advanced, which reduces the possibility of serious packet loss before the traffic is diverted.
  • 1 is a schematic diagram of a network topology of a hybrid access network
  • FIG. 2 is a flowchart of a method for offloading traffic according to an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for calculating a packet loss rate of a forwarding tunnel according to an embodiment of the present invention
  • FIG. 4 is a flowchart of a method for offloading traffic according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a method for offloading traffic according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of calculating a packet loss rate of a forwarding tunnel according to an embodiment of the present invention.
  • FIG. 7 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • FIG. 8 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • FIG. 9 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • FIG. 10 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • FIG. 11 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • FIG. 12 is a schematic structural diagram of hardware of a first network node according to an embodiment of the present invention.
  • the user terminal accesses the network by using the hybrid access mode
  • the forwarding traffic in the case that the forwarding traffic is small, the user terminal generally only establishes a forwarding tunnel on one of the data links A to forward the traffic, and only when the forwarding traffic exceeds the establishment.
  • the forwarding the rated bandwidth of the tunnel it will consider diverting the excess traffic to other data link B for forwarding.
  • the data link A is in various situations (for example, the bandwidth cannot be guaranteed and the number of access user terminals is large), the traffic that is forwarded in the forwarding tunnel may not reach the forwarding tunnel.
  • packet loss occurs.
  • the nominal bandwidth described herein refers to the bandwidth allocated by the data link A to the forwarding tunnel. It can be seen that the nominal bandwidth of the forwarding tunnel is less than or equal to the rated bandwidth of the data link A. That is to say, the system can not solve the packet loss phenomenon, and the existence of the packet loss phenomenon will greatly affect the user experience.
  • the first network node reduces the rated bandwidth of the first forwarding tunnel to the first available bandwidth when the packet loss ratio of the first forwarding tunnel is large, so that When the traffic forwarded by the first forwarding tunnel becomes larger, once the first available bandwidth is exceeded, the forwarding traffic exceeding the first available bandwidth is started to be offloaded to the second data link for forwarding, because The available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel, and the forwarding traffic will exceed the first available bandwidth earlier, and the splitting time of the first network node's traffic splitting is advanced, and the traffic is reduced before the traffic is split. The possibility of a serious packet loss situation increases, thereby improving the user experience.
  • the packet loss rate of the first forwarding tunnel When the packet loss rate of the first forwarding tunnel is gradually reduced to be less than the second preset threshold, it may be determined that the forwarding environment of the first forwarding tunnel can normally carry more traffic forwarding.
  • the first network device further increases the first available bandwidth of the first forwarding tunnel that has been reduced to a second predetermined bandwidth, so that the first network device can use the first forwarding tunnel to forward more than before.
  • the traffic of the second predetermined bandwidth so that the first forwarding tunnel carries more traffic forwarding, and improves the first forwarding bandwidth of the first forwarding tunnel when the first forwarding tunnel forwarding environment is improved.
  • the utilization of network resources When the packet loss rate of the first forwarding tunnel is gradually reduced to be less than the second preset threshold, it may be determined that the forwarding environment of the first forwarding tunnel can normally carry more traffic forwarding.
  • the first network device further increases the first available bandwidth of the first forwarding tunnel that has been reduced to a second predetermined bandwidth, so that the first network device can use the first forwarding tunnel to
  • FIG. 2 is a flowchart of a method for offloading traffic according to an embodiment of the present invention.
  • First network node and second network a first data link and a second data link are connected between the network nodes, and a first forwarding tunnel from the first network node to the second network node is established on the first data link,
  • the first network node and the second network node are respectively tunnel endpoints of the first forwarding tunnel.
  • the embodiment of the present invention can be applied to a scenario in which a hybrid access mode is used to access a network.
  • Two or more data links are respectively connected between two network nodes, where the first network node and the first network node
  • the first data link is used between the two network nodes as the priority data link for forwarding traffic, that is, the excess traffic is only generated when the traffic forwarded on the first data link exceeds a certain threshold.
  • the embodiment of the present invention does not limit that the user terminal is specifically connected to the first network node or the second network node.
  • the method includes:
  • the first network node calculates a first packet loss rate of the first forwarding tunnel in a preset period.
  • the first packet loss rate refers to the end-to-end packet loss rate of the first forwarding tunnel, that is, the data sent by the first network node to the second network node by using the first forwarding tunnel.
  • the number of packets is different from the number of data packets sent by the second network node to the first network node through the first forwarding tunnel.
  • the higher the packet loss rate the more the first forwarding tunnel cannot carry the current forwarding traffic.
  • An embodiment of the present invention provides a method for calculating a packet loss rate. As shown in FIG. 3, FIG. 3 is a flowchart of a method for calculating a packet loss rate of a forwarding tunnel according to an embodiment of the present invention, including:
  • the first network node sends a first counting packet to the second network node, and the first counting packet carries the first network node in the preset.
  • the total number of data packets is sent to the second network node by the first forwarding tunnel during the period.
  • the first network node receives the second counting packet that is returned by the second network node, where the second counting packet carries the first network node that is carried in the first counting packet. Transmitting a total number of data packets to the second network node by using the first forwarding tunnel in a preset period, and receiving, by the second network node, the first network by using the first forwarding tunnel in the preset period The total number of packets sent by the network node.
  • the first network node sends the total number of data packets to the second network node according to the first network node carried in the second counting packet, and the second network node receives the first network.
  • the total number of data packets sent by the node, and the first packet loss rate of the first forwarding tunnel corresponding to the preset period is calculated.
  • the preset period is 10 seconds, and the total number of data packets in the first count message sent by the first network node to the second network node is 1000, and the received second count message is received.
  • Receiving, by the second network node, the total number of data packets sent by the first network node is 950.
  • the total number of data packets in the first counting packet sent to the second node again is 1200, and the received
  • Receiving, by the node, the total number of data packets sent by the first network node is 1140, and the second packet returned by the two times may calculate that the first network node passes the first one in the preset period.
  • the forwarding tunnel sends 200 data packets.
  • the second network node receives 190 data packets through the first forwarding tunnel, and then 10 data packets are lost.
  • the packet loss rate can be expressed in 5% or other forms.
  • the average value of the packet loss rate of a plurality of preset periods may be calculated by using the total number of data packets of a plurality of preset periods, which is not limited by the present invention.
  • the first network node When the first packet loss rate is greater than the first predetermined threshold, the first network node reduces the rated bandwidth of the first forwarding tunnel by a first predetermined bandwidth to obtain a first available bandwidth.
  • the first network device determines whether the bandwidth condition for offloading to the second data link is the rated bandwidth of the first forwarding tunnel, and the embodiment of the present invention calculates the lost bandwidth.
  • the packet rate is reduced by the first predetermined bandwidth of the first forwarding tunnel to obtain a first available bandwidth, and the bandwidth condition for determining the offload is determined as the first available bandwidth.
  • the operation of triggering the traffic offloading may be triggered when the traffic forwarded in the first forwarding tunnel has not reached the rated bandwidth of the first forwarding tunnel but is greater than the first available bandwidth.
  • the first predetermined bandwidth may be decreased by the first predetermined bandwidth on the basis of the last decrease of the first predetermined bandwidth, until the first forwarding tunnel is caused.
  • the forwarded traffic reaches a new first available bandwidth to trigger traffic offloading or the packet loss rate in the first forwarding tunnel is less than the first predetermined threshold.
  • the first predetermined bandwidth it may be a preset fixed bandwidth, or may be a bandwidth determined according to a packet loss rate.
  • the present invention provides a method for calculating the first available bandwidth, as shown in Equation 1:
  • DL_TunnelBW(i) is the first available bandwidth of the first forwarding tunnel in the i period.
  • DL_BW is the nominal bandwidth of the first data link from the first network node to the second network node.
  • ⁇ DL_LBR(i) is the packet loss traffic corresponding to the packet loss rate accumulated in the first forwarding tunnel in the i period.
  • DL_BypassBW(i) is a bandwidth in the first data link of the i period from which the first network node to the second network node does not forward traffic through the first forwarding tunnel.
  • the first predetermined bandwidth of the first forwarding tunnel reduced by the rated bandwidth will be the first network device begins to calculate the packet loss rate.
  • Packet loss rate The total packet loss traffic.
  • the setting of the first predetermined bandwidth may be associated with the forwarding situation in the first forwarding tunnel, and the more serious the packet loss rate, the more bandwidth is reduced, so that the current location can be made faster.
  • the traffic forwarded on the first forwarding tunnel is greater than the first available bandwidth, and the first network device is triggered to perform traffic offloading.
  • Equation 1 also needs to consider that the first data link may have a part of traffic that is not forwarded by the first forwarding link from the first network node to the second network node, and this part of traffic The size will also change over time and forwarding.
  • the first network node reduces the rated bandwidth of the first forwarding tunnel to a first available bandwidth, so that the traffic that is forwarded by the first forwarding tunnel
  • the excess forwarding traffic is started to be offloaded to the second data link for forwarding. Since the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel, The forwarding traffic will exceed the first available bandwidth earlier, and the shunting time of the first network node to divert traffic is advanced, thereby reducing the possibility of serious packet loss before the traffic is split, thereby improving user usage.
  • the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel
  • the network forwarding state changes and the packet loss rate decreases.
  • the first network device needs to gradually restore the rated bandwidth of the first forwarding tunnel, and improve the ability of the first forwarding tunnel to forward traffic on the first data link to avoid waste of network resources.
  • the bandwidth of the first forwarding tunnel is increased by determining the packet loss rate of the first forwarding tunnel.
  • FIG. 4 is a flowchart of a method for offloading traffic according to an embodiment of the present invention, as shown in FIG. 4:
  • the first network node calculates a first packet loss rate of the first forwarding tunnel in a preset period.
  • the first network node When the first packet loss rate is greater than the first predetermined threshold, the first network node reduces the nominal bandwidth of the first forwarding tunnel by a first predetermined bandwidth to obtain a first available bandwidth.
  • the first network node increases the first available bandwidth by a second predetermined bandwidth to obtain a second available bandwidth, where the second predetermined bandwidth is less than or equal to The first predetermined bandwidth, the second available bandwidth is less than or equal to a nominal bandwidth of the first data link from the first network node to a second network node.
  • S404 does not need to be executed after traffic diversion in S403. If S402 is executed, the first available band is obtained. After the width, S404 may be performed to determine whether the first packet loss rate is less than the second predetermined threshold. The second predetermined threshold is less than the first predetermined threshold.
  • the first network device may incrementally increase the reduced first operating room bandwidth of the operating room without considering the first data link from the first network node to the second network node In the case of the traffic forwarded by the first forwarding tunnel, the increased second available bandwidth may not exceed the rated bandwidth of the first data link from the first network node to the second network node.
  • the bandwidth of the first forwarding tunnel may be continuously increased until the first packet loss rate.
  • the second predetermined threshold is again exceeded.
  • the first data link there is generally only a forwarding tunnel from the first network node to the second network node, and there may be a second network node to the first network node. Forward the tunnel.
  • FIG. 5 is a flowchart of a method for offloading traffic according to an embodiment of the present invention, where the first data link is further established from the second network node to the first network. a second forwarding tunnel of the node, where the first network node and the second network node are respectively tunnel endpoints of the second forwarding tunnel.
  • the method includes:
  • the first network node calculates a second packet loss rate of the second forwarding tunnel in the preset period.
  • the second forwarding tunnel forwards traffic from the second network node to the first network node, so when the first network node that is the receiver of the data packet is within a preset period of calculation
  • the packet loss rate of the second forwarding tunnel is used, the total number of data packets sent by the second network node by using the second forwarding tunnel is generally acquired.
  • the first network will be used. How the node calculates the second packet loss rate of the second forwarding tunnel is described.
  • the first network node When the second packet loss rate is greater than a third predetermined threshold, the first network node sends a bandwidth advertisement packet to the second network node, where the bandwidth advertisement packet is used to indicate the second network.
  • the node reduces the rated bandwidth of the second forwarding tunnel by a third predetermined bandwidth to obtain a third available bandwidth.
  • the first network node needs to indicate the second network, when the first network node is the destination endpoint of the second forwarding tunnel, when an operation of reducing the rated bandwidth of the second forwarding tunnel needs to be performed.
  • the node that is, the source endpoint of the second forwarding tunnel reduces the nominal bandwidth of the second forwarding tunnel to obtain the third available bandwidth.
  • the bandwidth advertisement packet may be a Generic Routing Encapsulation (GRE) packet.
  • the method further includes: when the forwarding traffic of the second forwarding tunnel exceeds the third available bandwidth, the second network node offloads the excess forwarding traffic to the second data. Forwarded on the link.
  • the technical solution of the present invention will be further described in the application scenario of accessing the network through a specific hybrid access mode.
  • the technical solution in this specific scenario is described based on the foregoing embodiments.
  • the first data link is a data link having a DSL link and an IP link
  • the second data link is an LTE link.
  • the first network node is an HG
  • the second network node is a HAAP
  • the first network node is a HAAP
  • the second network node is an HG.
  • the network topology of the application scenario is as shown in FIG.
  • the HG and the HAAP are established between a first forwarding tunnel from the HG to the HAAP and a second forwarding tunnel from the HAAP to the HG.
  • the first forwarding tunnel and the second forwarding tunnel are based on the first data link.
  • the first forwarding tunnel and the second forwarding tunnel may be GRE tunnels.
  • the HG calculates the first packet loss rate of the first forwarding tunnel, the HG may also calculate the second packet loss rate of the second forwarding tunnel at the same time.
  • FIG. 6 is a schematic diagram of calculating a packet loss rate of a forwarding tunnel according to an embodiment of the present invention.
  • the HG sends the first count message, where the first NUMBER message carries the total number of data packets (UL_Send) sent by the HG to the HAAP through the first forwarding tunnel.
  • the first NUMBER message carries the total number of data packets (UL_Send) sent by the HG to the HAAP through the first forwarding tunnel.
  • the HAAP After the first count message arrives at the HAAP, the HAAP adds the total number of data packets (UL_Recv) received by the HAAP through the first forwarding tunnel in the first count message, and at the same time, The HAAP may also add the total number of data packets (DL_Send) sent to the HG through the second forwarding tunnel to the first counting packet to form a second counting packet and pass the second forwarding tunnel. Sending to the HG, the second count message carries the UL_Send, UL_Recv, and DL_Send. After the second counting packet reaches the HG through the second forwarding tunnel, the HG may add the total number of data packets (DL_Recv) received through the second forwarding tunnel to the second counting packet.
  • DL_Recv total number of data packets
  • a third count message is formed and sent to the HAAP through the first forwarding tunnel.
  • the first count message, the second count message, and the third count message can all be extended on the basis of the existing GRE message. Since the GRE message includes a Type-length-value (TLV), the lengths of the T and L fields are often fixed (usually 1 to 4 bytes), and the length of the V field is variable.
  • TLV Type-length-value
  • the T field indicates the packet type
  • the L field indicates the packet length
  • the V field is used to store the content of the packet, so the scalability is very strong.
  • the HG sends the first count message every preset period, so that the HG can subtract the second count received last time by using the UL_Send carried in the second count message received this time.
  • Carryed in the message The UL_Send obtains the total number A of data packets sent to the HAAP through the first forwarding tunnel in the preset period between the previous time and the current time, and the HG may pass the second count message received this time.
  • the UL_Recv carried in the second received packet is received by the UL_Recv in the second received packet, and the packet received by the HAAP through the first forwarding tunnel in the preset period between the previous time and the current time is obtained.
  • subtracting B from A can obtain the first packet loss rate in the preset period.
  • the HG can also obtain the second packet loss rate of the second forwarding tunnel by using the DL_Send in the second counting packet and the DL_Recv obtained by itself.
  • the HG may adjust the bandwidth of the first forwarding tunnel by using Equation 1,
  • the DL_BW is 50M
  • the current DL_BypassBW is 5M
  • the current first available bandwidth is 40M of the DL_TunnelBW
  • the bandwidth occupied by the forwarding traffic in the first forwarding tunnel exceeds 40 M (the first available bandwidth)
  • the HG starts to offload traffic exceeding the 40 M bandwidth to the second data link (LTE link).
  • the traffic offloading is not required until the bandwidth occupied by the forwarding traffic in the first forwarding tunnel exceeds 45 M (the nominal bandwidth of the first forwarding tunnel).
  • the HG when the HG determines that the second packet loss rate is greater than a third predetermined threshold, the HG sends a bandwidth advertisement packet to the HAAP, where the bandwidth advertisement packet carries The value of the third available bandwidth after the third predetermined bandwidth is reduced, and the bandwidth advertisement message may be a Notify message.
  • the HAAP reduces the rated bandwidth of the second forwarding tunnel to the third available bandwidth according to the value of the third available bandwidth carried in the bandwidth advertisement packet.
  • the size of the second predetermined bandwidth may also be adjusted correspondingly according to different application scenarios.
  • the current network environment is a large number of user terminals connected to the network at night, and the number of user terminals connected to the network in the morning time is small.
  • the rated bandwidth of the forwarding tunnel is reduced by the first predetermined bandwidth by the technical solution of the present invention, and the packet loss rate in the forwarding tunnel is low.
  • the second predetermined bandwidth increased each time may select a smaller value that is much smaller than the first predetermined bandwidth. Since the night time is the peak of the Internet, the number of user terminals will remain at a high level.
  • the bandwidth needs to be reduced soon, resulting in the available bandwidth of the forwarding tunnel. Low, and this way of slowly increasing the bandwidth can be better adapted to the nighttime Internet environment, and there is no or less case where the available bandwidth of the forwarding tunnel is high or low. In the morning, the number of user terminals accessing the Internet is small. Even if there is a need to reduce the bandwidth of the forwarding tunnel, it should belong to the even discovery. Therefore, the size of the second predetermined bandwidth can be appropriately increased. After the technical solution of the present invention reduces the bandwidth of the forwarding tunnel, the rated bandwidth of the forwarding tunnel can be recovered quickly, thereby not wasting network resources.
  • the first network node reduces the rated bandwidth of the first forwarding tunnel to a first available bandwidth, so that the traffic that is forwarded by the first forwarding tunnel
  • the excess forwarding traffic is started to be offloaded to the second data link for forwarding. Since the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel, The forwarding traffic will exceed the first available bandwidth earlier, and the shunting time of the first network node to divert traffic is advanced, thereby reducing the possibility of serious packet loss before the traffic is split, thereby improving user usage.
  • the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel
  • the packet loss rate of the first forwarding tunnel is gradually reduced to be less than the second preset threshold, it may be determined that the forwarding environment of the first forwarding tunnel can normally carry more traffic forwarding.
  • the first network device further increases the first available bandwidth of the first forwarding tunnel that has been reduced by a second predetermined bandwidth, so that the first network device can use the first forwarding tunnel to forward the second reservation more than before.
  • the traffic of the bandwidth so that when the first forwarding tunnel forwarding environment is improved, the first forwarding tunnel carries more traffic forwarding by increasing the first available bandwidth of the first forwarding tunnel, thereby improving the network. Utilization of resources.
  • FIG. 7 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • a first data link and a second data link are connected between the first network node and the second network node, and the first data link is established with the first network node to the second network node a first forwarding tunnel, where the first network node and the second network node are tunnel endpoints of the first forwarding tunnel, respectively.
  • the split flow device 700 shown in FIG. 7 includes a first calculating unit 701, a bandwidth reducing unit 702, and a first shunting unit 703.
  • the first calculating unit 701 is configured to calculate a first packet loss rate of the first forwarding tunnel in a preset period.
  • the bandwidth reduction unit 702 is configured to reduce the rated bandwidth of the first forwarding tunnel by a first predetermined bandwidth when the execution result of the first calculating unit 701 is that the first packet loss rate is greater than a first predetermined threshold.
  • the first offloading unit 703 is configured to: when the forwarding traffic of the first forwarding tunnel exceeds the first available bandwidth, offload the excess forwarding traffic to the second data link for forwarding.
  • the first packet loss rate refers to the end-to-end packet loss rate of the first forwarding tunnel, that is, the data sent by the first network node to the second network node by using the first forwarding tunnel.
  • the number of packets is different from the number of data packets sent by the second network node to the first network node through the first forwarding tunnel.
  • the higher the packet loss rate the more the first forwarding tunnel cannot carry the current forwarding traffic.
  • Embodiments of the present invention provide an apparatus for calculating a packet loss rate. As shown in FIG. 8, FIG. 8 is a calculation and forwarding tunnel according to an embodiment of the present invention.
  • the device structure diagram of the channel packet loss rate, the first calculating unit 701 includes a first counting message sending subunit 801, a second counting message receiving subunit 802, and a first packet loss rate calculating subunit 803. among them:
  • a first counting message sending subunit 801 configured to send a first counting message to the second network node every the preset period, where the first counting message carries the first network node Sending, by the first forwarding tunnel, the total number of data packets to the second network node in the preset period.
  • a second count message receiving sub-unit 802 configured to receive a second count message returned by the second network node, where the second count message carries the first number carried in the first count message And the total number of data packets sent by the network node to the second network node by using the first forwarding tunnel in the preset period, and the second network node receiving by using the first forwarding tunnel in the preset period The total number of data packets sent by the first network node.
  • a first packet loss rate calculation unit 803 configured to send, according to the first network node that is carried in the second counting packet, the total number of data packets to the second network node, and the second network node receiving station. The total number of data packets sent by the first network node is calculated, and the first packet loss rate of the first forwarding tunnel corresponding to the preset period is calculated.
  • the preset period is 10 seconds
  • the total number of data packets in the first count message sent by the first counting message sending subunit 801 to the second network node is 1000
  • the second counting message is Receiving the returned second count message
  • the second network node receives the total number of data packets sent by the first network node by 950.
  • the first count message is sent.
  • the total number of data packets in the first count message sent by the subunit 801 to the second node is 1200
  • the second count message receiving subunit 802 receives the second in the returned second count message.
  • the network node receives the total number of data packets sent by the first network node as 1140, and the first packet loss rate calculation sub-unit 803 can calculate, by using the two returned second packets, that the preset period is
  • the first network node sends a total of 200 data packets through the first forwarding tunnel, and the second network node receives 190 data packets through the first forwarding tunnel, and then 10 data packets are lost.
  • Package rate can be 5% or other Formal representation.
  • the average value of the packet loss rate of a plurality of preset periods may be calculated by using the total number of data packets of a plurality of preset periods, which is not limited by the present invention.
  • the first network device determines whether the bandwidth condition for offloading to the second data link is the rated bandwidth of the first forwarding tunnel, and the embodiment of the present invention calculates the lost bandwidth. a packet rate, reducing a nominal bandwidth of the first forwarding tunnel by a first predetermined bandwidth, obtaining a first available bandwidth, and determining a bandwidth condition of the offload to determine each of the first available bandwidth, so that the first forwarding tunnel may be in the first forwarding tunnel.
  • the operation of the traffic splitting is triggered when the forwarded traffic has not reached the rated bandwidth of the first forwarding tunnel but is greater than the first available bandwidth.
  • the first available bandwidth may be decreased by the first predetermined bandwidth on the basis of the last reduction of the first predetermined bandwidth until the first The traffic forwarded in the forwarding tunnel reaches the first available bandwidth to trigger traffic offloading or the packet loss rate in the first forwarding tunnel is less than the first predetermined threshold.
  • the first predetermined bandwidth it may be a preset fixed bandwidth, or may be a bandwidth determined according to a packet loss rate.
  • the present invention is optional.
  • a way of calculating the first available bandwidth is provided, as shown in Equation 1.
  • the first network node reduces the rated bandwidth of the first forwarding tunnel to a first available bandwidth, so that the traffic that is forwarded by the first forwarding tunnel
  • the excess forwarding traffic is started to be offloaded to the second data link for forwarding. Since the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel, The forwarding traffic will exceed the first available bandwidth earlier, and the shunting time of the first network node to divert traffic is advanced, thereby reducing the possibility of serious packet loss before the traffic is split, thereby improving user usage.
  • the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel
  • the network forwarding state changes and the packet loss rate decreases.
  • the first network device needs to gradually restore the rated bandwidth of the first forwarding tunnel, and improve the ability of the first forwarding tunnel to forward traffic on the first data link to avoid waste of network resources.
  • the bandwidth of the first forwarding tunnel is increased by determining the packet loss rate of the first forwarding tunnel.
  • FIG. 9 is a structural diagram of a device for distributing traffic according to an embodiment of the present invention.
  • the traffic distribution device 700 further includes a bandwidth increasing unit 901.
  • the adding unit 901 is configured to: when the execution result of the first calculating unit 701 is that the first packet loss rate is less than a second preset threshold, increase the first available bandwidth by a second predetermined bandwidth, to obtain a second available The bandwidth, the second predetermined bandwidth is less than or equal to the first predetermined bandwidth, and the second available bandwidth is less than or equal to a nominal bandwidth of the first data link from the first network node to the second network node.
  • the first shunting unit 703 and the bandwidth increasing unit 901 do not define a triggering relationship, and the bandwidth increasing unit 901 does not need to trigger the first shunting unit 703.
  • the bandwidth increase unit 901 can be triggered to determine whether the first packet loss rate is smaller than the second preset threshold.
  • the second preset threshold is smaller than the first preset threshold.
  • the first The network device may incrementally increase the reduced first operating room bandwidth of the operating room without passing through the first forwarding from the first network node to the second network node regardless of the first data link In the case of tunnel-forwarded traffic, the increased second available bandwidth may not exceed the nominal bandwidth of the first data link from the first network node to the second network node.
  • the bandwidth of the first forwarding tunnel may be continuously increased until the first packet loss. The rate again exceeds the second predetermined threshold.
  • the first data link there is generally only a forwarding tunnel from the first network node to the second network node, and there may be a second network node to the first network node. Forward the tunnel.
  • FIG. 10 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • a second forwarding tunnel from the second network node to the first network node is further established in the first data link, where the first network node and the second network node are the second forwarding tunnel respectively Tunnel endpoint.
  • the split flow device 700 further includes a second calculating unit 1001 and a sending unit 1002, where the second calculating unit 1001 is configured to calculate a second forwarding tunnel in the preset period. Two packet loss rate.
  • the sending unit 1002 is configured to send a bandwidth advertisement message to the second network node, where the second packet loss rate is greater than a third predetermined threshold, when the execution result of the second calculation unit is greater than a third predetermined threshold, the bandwidth advertisement packet And configured to instruct the second network node to reduce a nominal bandwidth of the second forwarding tunnel by a third predetermined bandwidth to obtain a third available bandwidth.
  • the second forwarding tunnel forwards traffic from the second network node to the first network node, so when the first network node that is the received data packet calculates the preset period
  • the packet loss rate of the second forwarding tunnel is described, the total number of data packets sent by the second network node by using the second forwarding tunnel is generally obtained.
  • the second computing unit is used. 1001 describes how to calculate the second packet loss rate of the second forwarding tunnel.
  • the first network node needs to indicate the second network, when the first network node is the destination endpoint of the second forwarding tunnel, when an operation of reducing the rated bandwidth of the second forwarding tunnel needs to be performed.
  • the node that is, the source endpoint of the second forwarding tunnel reduces the nominal bandwidth of the second forwarding tunnel to obtain the third available bandwidth.
  • the bandwidth advertisement packet may be a Generic Routing Encapsulation (GRE) packet.
  • FIG. 11 is a structural diagram of a device for splitting traffic according to an embodiment of the present invention.
  • the split flow device 1100 includes:
  • the second offloading unit 1101 is configured to: when the forwarding traffic of the second forwarding tunnel exceeds the third available bandwidth, offload the excess forwarding traffic to the second data link for forwarding.
  • the first data link is a digital subscriber line DSL link and a network protocol IP link.
  • the second data link is a Long Term Evolution (LTE) link.
  • LTE Long Term Evolution
  • the second network node is a hybrid access aggregation point HAAP.
  • the second network node is an HG.
  • the first network node reduces the rated bandwidth of the first forwarding tunnel to a first available bandwidth, so that the traffic that is forwarded by the first forwarding tunnel
  • the excess forwarding traffic is started to be offloaded to the second data link for forwarding. Since the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel, The forwarding traffic will exceed the first available bandwidth earlier, and the shunting time of the first network node to divert traffic is advanced, thereby reducing the possibility of serious packet loss before the traffic is split, thereby improving user usage.
  • the first available bandwidth is smaller than the original rated bandwidth of the first forwarding tunnel
  • the packet loss rate of the first forwarding tunnel is gradually reduced to be less than the second preset threshold, it may be determined that the forwarding environment of the first forwarding tunnel can normally carry more traffic forwarding.
  • the first network device further increases the first available bandwidth of the first forwarding tunnel that has been reduced by a second predetermined bandwidth, so that the first network device can use the first forwarding tunnel to forward the second reservation more than before.
  • the traffic of the bandwidth so that when the first forwarding tunnel forwarding environment is improved, the first forwarding tunnel carries more traffic forwarding by increasing the first available bandwidth of the first forwarding tunnel, thereby improving the network. Utilization of resources.
  • FIG. 12 is a schematic diagram of a hardware structure of a first network node according to an embodiment of the present invention, where a first data link and a second data link are connected between the first network node 1200 and a second network node.
  • a first forwarding tunnel from the first network node 1200 to the second network node, where the first network node 1200 and the second network node are respectively
  • the tunnel endpoint of the first forwarding tunnel the first network node 1200 includes a memory 1201 and a processor 1202 coupled to the memory 1201, the memory 1201 is configured to store a set of program instructions, and the processor 1202 is configured to The program instructions stored in the memory 1201 are called to perform the following operations:
  • the excess forwarding traffic is offloaded to the second data link for forwarding.
  • the processor 1202 may be a central processing unit (CPU), and the memory 1201 may be an internal memory of a random access memory (RAM) type.
  • the processor 1202 and the memory 1201 may be integrated into one or more independent circuits or hardware, such as an Application Specific Integrated Circuit (ASIC).
  • ASIC Application Specific Integrated Circuit
  • the first network node, the first data link, the first forwarding tunnel, the first packet loss rate, the first predetermined threshold, the first predetermined bandwidth, the first available bandwidth, and the first count packet mentioned in the embodiment of the present invention The "first” is only used to name the name, not the first in the order. The same rules apply to "second" and "third".
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

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

Abstract

本发明实施例公开了一种分流流量的方法和装置，包括：第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率；当所述第一丢包率大于第一预定阈值时，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；当所述第一转发隧道的转发流量超出所述第一可用带宽时，将超出的转发流量分流到所述第二数据链路上转发，可以看出，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的可用带宽为第一可用带宽，由于所述第一可用带宽小于原本的所示第一转发隧道的可用带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性。

Description

一种分流流量的方法和装置
本申请要求于2014年8月05日提交中国专利局、申请号为201410382683.1、发明名称为“一种分流流量的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及通信领域，特别是涉及一种分流流量的方法和装置。
背景技术
混合接入(Hybrid Access，HA)是目前是比较常用的用户终端连接网络的方式，所谓的混合接入方式是指用户终端同时通过多种不同类型的数据传输链路与网络相连。混合接入方式中比较典型的应用场景是用户终端同时通过数字用户线路(Digital Subscriber Line，DSL)链路和长期演进(Long Term Evolution，LTE)链路与网络相连。
图1为混合接入网络的网络拓扑示意图，如图1所示，PC(个人电脑)作为用户终端，与家庭网关(Home Gateway，HG)相连，所述HG和混合接入汇聚点(Hybrid Access Aggregation Point，HAAP)之间有DSL链路和LTE链路，所述HAAP为接入网络的网络节点。比如说DSL链路的带宽为50M，LTE链路的带宽为100M，用户终端首先使用DSL链路与网络进行数据交换，当用户终端与网络之间的流量超出DSL链路能够承载的50M时，出现溢出现象(overflow)，超出的DSL链路带宽的流量将被分流到LTE链路上进行转发。
然而，其中HG1和HG2与数字用户线路接入复用器(Digital Subscriber Line Access Multiplexer，DSLAM)之间的DSL链路上的带宽是可以保证的，但是所述DSLAM通过路由器(Router)到HAAP之间是IP链路，提供给每一个用户终端的带宽无法保证，由此导致当所述DSLAM接入的HG数量过多时，即使DSL链路上的流量还没有超出DSL链路带宽，但在IP链路上转发流量的过程中依然会出现比较严重的丢包现象，影响了网络质量。
发明内容
本发明提供了一种分流流量的方法和装置，以减少丢包。
第一方面，本发明实施例提供了一种分流流量的方法，第一网络节点和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点，包括：
所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率；
当所述第一丢包率大于第一预定阈值时，所述第一网络节点将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；
当所述第一转发隧道的转发流量超出所述第一可用带宽时，所述第一网络节点将超出的转发流量分流到所述第二数据链路上转发。
在第一方面的第一种可能的实现方式中，该方法还包括：
当所述第一丢包率小于第二预设阈值时，所述第一网络节点将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于等于所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
结合第一方面或者第一方面的第一种可能的实现方式，在第二种可能的实现方式中，所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率，具体包括：
每隔所述预设周期，所述第一网络节点向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数；
所述第一网络节点接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数；
所述第一网络节点根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
在第一方面的第三种可能的实现方式中，所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分 别是所述第二转发隧道的隧道端点，该方法包括：
所述第一网络节点计算在所述预设周期内所述第二转发隧道的第二丢包率；
当所述第二丢包率大于第三预定阈值时，所述第一网络节点向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
结合第一方面的第三种可能的实现方式，在第四种可能的实现方式中，该方法还包括：
当所述第二转发隧道的转发流量超出所述第三可用带宽时，所述第二网络节点将超出的转发流量分流到所述第二数据链路上转发。
结合第一方面或者第一方面的第一种或第二种或第三种或第四种可能的实现方式，在第五种可能的实现方式中，
所述第一数据链路为具有数字用户线路DSL链路和网络协议IP链路；
所述第二数据链路为长期演进LTE链路。
结合第一方面或者第一方面的第一种或第二种或第三种或第四种或第五种可能的实现方式，在第六种可能的实现方式中，
当所述第一网络节点为家庭网关HG时，所述第二网络节点为混合接入汇聚点HAAP；
当所述第一网络节点为HAAP时，所述第二网络节点为HG。
第二方面，本发明实施例提供了一种分流流量的装置，第一网络节点和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点，包括：
第一计算单元，用于计算在预设周期内所述第一转发隧道的第一丢包率；
带宽减少单元，用于当所述第一计算单元的执行结果为所述第一丢包率大于第一预定阈值时，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；
第一分流单元，用于当所述第一转发隧道的转发流量超出所述第一可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
在第二方面的第一种可能的实现方式中，该装置还包括：
带宽增加单元，用于当所述第一计算单元的执行结果为所述第一丢包率小于第二预设阈值时，将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于等于所述第一数据链路中从 所述第一网络节点向第二网络节点的额定带宽。
结合第二方面或者第二方面的第一种可能的实现方式，在第二种可能的实现方式中，所述第一计算单元，包括：
第一计数报文发送子单元，用于每隔所述预设周期，向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数；
第二计数报文接收子单元，用于接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数；
第一丢包率计算子单元，用于根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
在第二方面的第三种可能的实现方式中，所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分别是所述第二转发隧道的隧道端点，该装置包括：
第二计算单元，用于计算在所述预设周期内所述第二转发隧道的第二丢包率；
发送单元，用于当所述第二计算单元的执行结果为所述第二丢包率大于第三预定阈值时，向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
结合第二方面的第三种可能的实现方式，在第四种可能的实现方式中，该装置还包括：
第二分流单元，用于当所述第二转发隧道的转发流量超出所述第三可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
结合第二方面或者第二方面的第一种或第二种或第三种或第四种可能的实现方式，在第五种可能的实现方式中，
所述第一数据链路为具有数字用户线路DSL链路和网络协议IP链路；
所述第二数据链路为长期演进LTE链路。
结合第二方面或者第二方面的第一种或第二种或第三种或第四种或第五种可能的实现方式，在第六种可能的实现方式中，
当所述第一网络节点为家庭网关HG时，所述第二网络节点为混合接入汇聚点HAAP；
当所述第一网络节点为HAAP时，所述第二网络节点为HG。
由上述技术方案可以看出，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的可用带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所示第一转发隧道的可用带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为混合接入网络的网络拓扑示意图；
图2为本发明实施例提供的一种分流流量的方法流程图；
图3为本发明实施例提供的一种计算转发隧道丢包率的方法流程图；
图4为本发明实施例提供的一种分流流量的方法流程图；
图5为本发明实施例提供的一种分流流量的方法流程图；
图6为本发明实施例提供的一种计算转发隧道丢包率的示意图；
图7为本发明实施例提供的一种分流流量的装置结构图；
图8为本发明实施例提供的一种分流流量的装置结构图；
图9为本发明实施例提供的一种分流流量的装置结构图；
图10为本发明实施例提供的一种分流流量的装置结构图；
图11为本发明实施例提供的一种分流流量的装置结构图；
图12为本发明实施例提供的一种第一网络节点的硬件结构示意图。
具体实施方式
在用户终端使用混合接入的方式接入网络的情况下，在转发流量较小的情况下，用户终端一般只在其中一条数据链路A上建立转发隧道来转发流量，只有当转发流量超出建立的转发隧道的额定带宽时，才会考虑将超出部分的流量分流到其他数据链路B上转发。但是如果所述数据链路A由于各种情况(比如不能保证带宽且接入用户终端数量较多的情况下)，可能会出现在所述转发隧道中转发的流量还没有达到所述转发隧道的额定带宽时，就出现了丢包的现象，但是由于转发流量还未超过所述转发隧道的额定带宽，系统也不会将所述转发流量分流到其他数据链路B上进行转发。这里所述的额定带宽是指所述数据链路A为所述转发隧道分配的带宽大小，可以看出，所述转发隧道的额定带宽小于等于所述数据链路A的额定带宽。也就是说，系统面对出现的丢包现象是无法进行解决的，由此丢包现象一直存在将大大影响用户的使用体验。为此，在本发明提供的实施例中，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的额定带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出所述第一可用带宽的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所述第一转发隧道的额定带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性，由此提高了用户的使用体验。
当第一转发隧道上的丢包率逐渐减小到小于第二预设阈值时，则可以判定第一转发隧道的转发环境可以正常承载更多的流量转发。所述第一网络设备将已经减少后的第一转发隧道的第一可用带宽进一步增加到第二预定带宽，以使得所述第一网络设备可以使用所述第一转发隧道转发比之前多的第二预定带宽的流量，由此可以在所述第一转发隧道转发环境改善时，通过提高所述第一转发隧道的第一可用带宽，让所述第一转发隧道承载较多的流量转发，提高了网络资源的利用率。
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
实施例一
图2为本发明实施例提供的一种分流流量的方法流程图。第一网络节点和第二网 络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点。
可以看出本发明实施例可以应用在使用混合接入方式接入网络的场景中，在两个网络节点之间分别连接有两条甚至多条数据链路，在所述第一网络节点和第二网络节点之间使用所述第一数据链路作为转发流量的优先数据链路，也就是说，只有在所述第一数据链路上转发的流量超出一定阈值时，才会将超出的流量分流到第二数据链路上转发。在具有用户终端的应用场景中，本发明实施例并不限定用户终端具体与所述第一网络节点或所述第二网络节点相连。
如图2所示，所述方法包括：
S201：所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率。
这里需要说明的是，所述第一丢包率是指所述第一转发隧道端到端的丢包率，即所述第一网络节点通过所述第一转发隧道向第二网络节点发出的数据包数量与所述第二网络节点通过所述第一转发隧道接收到所述第一网络节点发送的数据包数量的差值。丢包率越高，表明所述第一转发隧道越无法正常承载当前的转发流量。本发明实施例提供了一种计算丢包率的方法，如图3所示，图3为本发明实施例提供的一种计算转发隧道丢包率的方法流程图，包括：
S301：每隔所述预设周期，所述第一网络节点向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数。
S302：所述第一网络节点接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数。
S303：所述第一网络节点根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
比如说，预设周期为10秒，所述第一网络节点向所述第二网络节点发送的第一计数报文中的数据包总数为1000，接收到返回的第二计数报文中所述第二网络节点接收所述第一网络节点发送的数据包总数为950。10秒预设周期之后，再次向所述第二节点发送的第一计数报文中的数据包总数为1200，接收到返回的第二计数报文中所述第二网络 节点接收所述第一网络节点发送的数据包总数为1140，那么通过这两次返回的第二报文可以计算得出在所述预设周期内所述第一网络节点共通过所述第一转发隧道发送200个数据包，所述第二网络节点共通过所述第一转发隧道接收了190个数据包，那么就丢了10数据包，丢包率可以以5％或其他形式表示。当然也可以通过多个预设周期的数据包总数来计算多个预设周期的丢包率平均值等，本发明对此不进行限定。
S202：当所述第一丢包率大于第一预定阈值时，所述第一网络节点将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽。
这里需要说明的是，原本现有技术中所述第一网络设备判断是否向所述第二数据链路分流的带宽条件为所述第一转发隧道的额定带宽，而本发明实施例通过计算丢包率，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽，并将判断分流的带宽条件确定为所述第一可用带宽。这样可以当在所述第一转发隧道中转发的流量还未达到所述第一转发隧道的额定带宽，却已经大于所述第一可用带宽时就触发流量分流的操作。还需要注意的是，如果减少一次第一预定带宽后，所述第一转发隧道中转发的流量依然没有达到所述第一可用带宽，没法触发流量分流，但是所述第一转发隧道中的丢包率仍然大于所述第一预定阈值时，可以在上一次减少第一预定带宽的基础上，继续将所述第一可用带宽减少所述第一预定带宽，直到使得所述第一转发隧道中转发的流量达到新的第一可用带宽触发流量分流或者所述第一转发隧道中的丢包率小于所述第一预定阈值。
对于所述第一预定带宽，可以是预设的固定带宽，也可以是根据丢包率确定的带宽。当所述第一预定带宽根据丢包率确定的带宽时，可选的，本发明提供了一种计算第一可用带宽的方式，如公式1所示：
DL_TunnelBW(i)＝DL_BW-∑DL_LBR(i)-DL_BypassBW(i)  (公式1)
其中：
DL_TunnelBW(i)为i周期所述第一转发隧道的第一可用带宽。
DL_BW为所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
∑DL_LBR(i)为i周期所述第一转发隧道中累积的丢包率对应的丢包流量。
DL_BypassBW(i)为i周期所述第一数据链路中从所述第一网络节点向第二网络节点不通过所述第一转发隧道转发流量的带宽。
通过公式1可以看出，当丢包率大于所述第一预定阈值时，所述第一转发隧道的额定带宽减少的第一预定带宽将是所述第一网络设备开始计算丢包率以来的丢包率的 总和对应的丢包流量。这种对所述第一预定带宽的设定可以与所述第一转发隧道中的转发情况建立对应关系，丢包率越严重，减少的带宽就越多，这样可以更快的使得当前在所述第一转发隧道上转发的流量大于第一可用带宽，触发所述第一网络设备进行流量分流。同时可以看出，公式1还需要考虑到所述第一数据链路从所述第一网络节点向第二网络节点中可能具有一部分不通过所述第一转发链路转发的流量，这部分流量的大小也会随着时间以及转发情况改变。
S203：当所述第一转发隧道的转发流量超出所述第一可用带宽时，所述第一网络节点将超出的转发流量分流到所述第二数据链路上转发。
可见，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的额定带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所示第一转发隧道的额定带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性，由此提高了用户的使用体验。
当所述第一转发隧道由于丢包率较大，将所述第一转发隧道的额定带宽减少了所述第一预订带宽后，随着网络转发状态的改变以及丢包率的降低，所述第一网络设备需要将逐步恢复所述第一转发隧道的额定带宽，提高所述第一数据链路上的所述第一转发隧道转发流量的能力，以避免网络资源的浪费。本发明实施例通过对所述第一转发隧道丢包率大小的判断，对所述第一转发隧道的带宽进行增加。
图4为本发明实施例提供的一种分流流量的方法流程图，如图4所示：
S401：所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率。
S402：当所述第一丢包率大于第一预定阈值时，所述第一网络节点将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽。
S403：当所述第一转发隧道的转发流量超出所述第一可用带宽时，所述第一网络节点将超出的转发流量分流到所述第二数据链路上转发。
S404：当所述第一丢包率小于第二预定阈值时，所述第一网络节点将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于等于所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
首先需要注意的是，S403和S404之间并没有限定步骤执行的先后关系，S404并不需要一定要在S403进行流量分流之后才能执行，只要执行了S402，得到第一可用带 宽后即可以执行S404，判断所示第一丢包率是否小于所述第二预定阈值。所述第二预定阈值小于所述第一预定阈值。
当所述第一丢包率小于所述第二预定阈值时，则代表当前所述第一转发链路的转发情况较好(比如说当前连接所述第一转发隧道的用户终端的数量减少等)，所述第一网络设备可以将减少的手术室第一预设带宽的逐步的增加回来，在不考虑所述第一数据链路中从所述第一网络节点向第二网络节点中没有通过所述第一转发隧道转发的流量的情况下，增加后的第二可用带宽最大不能超过所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
当增加一次所述第二预订带宽后，如果所述第一丢包率持续小于所述第二预定阈值，可以持续对所述第一转发隧道的带宽进行增加，直到所述第一丢包率再一次超过所述第二预定阈值。
在所述第一数据链路中，一般不会只有从所述第一网络节点到所述第二网络节点的转发隧道，还可能有从所述第二网络节点到所述第一网络节点的转发隧道。
可选的，基于图4的方法，图5为本发明实施例提供的一种分流流量的方法流程图，所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分别是所述第二转发隧道的隧道端点。如图5所示，该方法包括：
S501：所述第一网络节点计算在所述预设周期内所述第二转发隧道的第二丢包率。
需要注意的是，所述第二转发隧道是从所述第二网络节点向所述第一网络节点转发流量的，故当作为数据包接收方的所述第一网络节点在计算预设周期内所述第二转发隧道的丢包率时，一般需要首先获取所述第二网络节点通过所述第二转发隧道发送的数据包总数，在下面的具体应用场景中，将对所述第一网络节点如何计算所述第二转发隧道的第二丢包率进行描述。
S502：当所述第二丢包率大于第三预定阈值时，所述第一网络节点向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
由于所述第一网络节点为所述第二转发隧道的目的端点，故当需要进行将所述第二转发隧道的额定带宽减少的操作时，所述第一网络节点需要指示所述第二网络节点，也就是所述第二转发隧道的源端点减少所述第二转发隧道的额定带宽，得到所述第三可用带宽。所述带宽通告报文可以是通用路由封装(Generic Routing Encapsulation，GRE)报文。
当所述第二转发隧道上转发的流量超出了所述第三可用带宽时，执行对超出部分的带宽进行分流操作的也是所述第二网络节点。也就是说，可选的，该方法还包括：当所述第二转发隧道的转发流量超出所述第三可用带宽时，所述第二网络节点将超出的转发流量分流到所述第二数据链路上转发。
接下来将通过具体的混合接入方式接入网络的应用场景对本发明的技术方案进行进一步的描述。该具体场景下的技术方案，基于前述实施例来描述。所述第一数据链路为具有DSL链路和IP链路的数据链路，所述第二数据链路为LTE链路。当所述第一网络节点为HG时，所述第二网络节点为HAAP；当所述第一网络节点为HAAP时，所述第二网络节点为HG。在本应用场景中，假定所述第一网络节点为HG，所述第二网络节点为HAAP，本应用场景的网络拓扑如所述图1所示，所述HG和所述HAAP之间建立有从所述HG到所述HAAP的第一转发隧道和从所述HAAP到所述HG的第二转发隧道。所述第一转发隧道和所述第二转发隧道基于所述第一数据链路。可选的，所述第一转发隧道和所述第二转发隧道可以是GRE隧道。所述HG在计算所述第一转发隧道的所述第一丢包率时，可选的，所述HG还可以同时计算所述第二转发隧道的所述第二丢包率。
如图6所示，图6为本发明实施例提供的一种计算转发隧道丢包率的示意图。在图6中，所述HG发送所述第一计数报文，所述第一记数报文中携带有所述HG向所述HAAP通过所述第一转发隧道发送的数据包总数(UL_Send)，当所述第一计数报文到达所述HAAP后，所述HAAP在所述第一计数报文中添加所述HAAP通过所述第一转发隧道接收的数据包总数(UL_Recv)，同时，所述HAAP还可以将通过所述第二转发隧道向所述HG发送的数据包总数(DL_Send)也添加到所述第一计数报文中，形成第二计数报文并通过所述第二转发隧道向所述HG发送，所述第二计数报文中携带有所述UL_Send、UL_Recv和DL_Send。当所述第二计数报文通过所述第二转发隧道到达所述HG后，所述HG可以将通过所述第二转发隧道接收的数据包总数(DL_Recv)添加到所述第二计数报文中，形成第三计数报文并通过所述第一转发隧道向所述HAAP发送。所述第一计数报文、第二计数报文和第三计数报文均可以在现有的GRE报文的基础上进行扩展得到。由于所述GRE报文包括类型-长度-值(Type-length-value，TLV)，T、L字段的长度往往固定(通常为1～4bytes)，V字段长度可变。T字段表示报文类型，L字段表示报文长度、V字段往往用来存放报文的内容，所以扩展性很强。
所述HG每隔预设周期发送所述第一计数报文，由此所述HG可以通过本次接收的所述第二计数报文中携带的UL_Send减去上一次接收的所述第二计数报文中携带的 UL_Send得到上一次到本次之间的所述预设周期中通过所述第一转发隧道向所述HAAP发送的数据包总数A，所述HG可以通过本次接收的所述第二计数报文中携带的UL_Recv减去上一次接收的所述第二计数报文中携带的UL_Recv得到上一次到本次之间的所述预设周期中所述HAAP通过所述第一转发隧道接收的数据包总数B，将A减去B便可以得到所述预设周期内的所述第一丢包率。同理，所述HG也可以通过所述第二计数报文中的DL_Send以及自身获取的DL_Recv得到所述第二转发隧道的第二丢包率。
以第一转发隧道为例，当所述HG确定所述第一丢包率大于第一预设阈值时，所述HG可以通过所述公式1对所述第一转发隧道的带宽进行调整，假设DL_BW为50M，当前的DL_BypassBW为5M，∑DL_LBR为5M时，当前的所述第一可用带宽也就是所述DL_TunnelBW为40M，而所述第一转发隧道的额定带宽为50M-5M＝45M。当所述第一转发隧道中的转发流量所占用的带宽超过40M(第一可用带宽)时，所述HG就开始将超出40M带宽的流量分流到所述第二数据链路(LTE链路)中，而不再需要继续等到所述第一转发隧道中的转发流量所占用的带宽超过45M(所述第一转发隧道的额定带宽)才进行流量分流。
以第二转发隧道为例，当所述HG确定所述第二丢包率大于第三预定阈值时，所述HG将向所述HAAP发送带宽通告报文，所述带宽通告报文中携带有减少第三预定带宽后的第三可用带宽的值，所述带宽通告报文可以是通告(Notify)报文。所述HAAP根据所述带宽通告报文中携带的第三可用带宽的值减少所述第二转发隧道的额定带宽为所述第三可用带宽。
所述第二预定带宽的大小还可以根据不同的应用场景进行对应的调整。比如说目前的网络环境都是晚上的时段连接网络的用户终端数量较多，而早上的时段连接网络的用户终端数量较少，针对这种典型的网络使用情况，在晚上的时段，当转发隧道转发的流量还未超出额定带宽且丢包率较高时，通过本发明的技术方案将所述转发隧道的额定带宽减少所述第一预定带宽，并在所述转发隧道的丢包率较低时，每次增加的所述第二预定带宽可以选择一个远小于所述第一预定带宽的较小的值。由于晚上时段属于上网高峰，用户终端的数量会保持在较多的程度，如果一次增加较多带宽的话，可能会很快又出现需要减少带宽的情况，由此导致转发隧道的可用带宽忽高忽低，而这种缓慢增加带宽的方式可以比较好的适应晚上的上网环境，不会或较少出现转发隧道的可用带宽忽高忽低的情况。而早上时段上网的用户终端数量较少，即使出现了需要减少转发隧道带宽的情况，也应该属于偶发现象，故可以适当调高所述第二预定带宽的大小，这样在通过 本发明的技术方案减少了转发隧道的带宽后，可以较快的恢复所述转发隧道的额定带宽，由此不会造成网络资源的浪费。
可见，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的额定带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所示第一转发隧道的额定带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性，由此提高了用户的使用体验。当第一转发隧道上的丢包率逐渐减小到小于第二预设阈值时，则可以判定第一转发隧道的转发环境可以正常承载更多的流量转发。所述第一网络设备将已经减少后的第一转发隧道的第一可用带宽进一步增加第二预定带宽，以使得所述第一网络设备可以使用所述第一转发隧道比之前多转发第二预定带宽的流量，由此可以在所述第一转发隧道转发环境改善时，通过提高所述第一转发隧道的第一可用带宽，让所述第一转发隧道承载较多的流量转发，提高了网络资源的利用率。
实施例二
图7为本发明实施例提供的一种分流流量的装置结构图。第一网络节点和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点。图7所示的分流流量装置700包括第一计算单元701、带宽减少单元702和第一分流单元703。其中，第一计算单元701，用于计算在预设周期内所述第一转发隧道的第一丢包率。带宽减少单元702，用于当所述第一计算单元701的执行结果为所述第一丢包率大于第一预定阈值时，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽。第一分流单元703，用于当所述第一转发隧道的转发流量超出所述第一可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
这里需要说明的是，所述第一丢包率是指所述第一转发隧道端到端的丢包率，即所述第一网络节点通过所述第一转发隧道向第二网络节点发出的数据包数量与所述第二网络节点通过所述第一转发隧道接收到所述第一网络节点发送的数据包数量的差值。丢包率越高，表明所述第一转发隧道越无法正常承载当前的转发流量。本发明实施例提供了一种计算丢包率的装置。如图8所示，图8为本发明实施例提供的一种计算转发隧 道丢包率的装置结构图，所述第一计算单元701包括第一计数报文发送子单元801、第二计数报文接收子单元802和第一丢包率计算子单元803。其中：
第一计数报文发送子单元801，用于每隔所述预设周期，向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数。
第二计数报文接收子单元802，用于接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数。
第一丢包率计算子单元803，用于根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
比如说，预设周期为10秒，所述第一计数报文发送子单元801向所述第二网络节点发送的第一计数报文中的数据包总数为1000，所述第二计数报文接收子单元802接收到返回的第二计数报文中所述第二网络节点接收所述第一网络节点发送的数据包总数为950。10秒预设周期之后，所述第一计数报文发送子单元801再次向所述第二节点发送的第一计数报文中的数据包总数为1200，所述第二计数报文接收子单元802接收到返回的第二计数报文中所述第二网络节点接收所述第一网络节点发送的数据包总数为1140，那么所述第一丢包率计算子单元803通过这两次返回的第二报文可以计算得出在所述预设周期内所述第一网络节点共通过所述第一转发隧道发送200个数据包，所述第二网络节点共通过所述第一转发隧道接收了190个数据包，那么就丢了10数据包，丢包率可以以5％或其他形式表示。当然也可以通过多个预设周期的数据包总数来计算多个预设周期的丢包率平均值等，本发明对此不进行限定。
这里需要说明的是，原本现有技术中所述第一网络设备判断是否向所述第二数据链路分流的带宽条件为所述第一转发隧道的额定带宽，而本发明实施例通过计算丢包率，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽，并将判断分流的带宽条件确定各位所述第一可用带宽，这样可以在所述第一转发隧道中转发的流量还未达到所述第一转发隧道的额定带宽，却已经大于所述第一可用带宽时就触发流量分流的操作。还需要注意的是，如果减少一次第一预定带宽后，所述第一转发隧道中转发的流量依然没有达到所述第一可用带宽，没法触发流量分流，但是所述第一转发隧道 中的丢包率仍然大于所述第一预定阈值时，可以在上一次减少第一预定带宽的基础上，继续将所述第一可用带宽减少所述第一预定带宽，直到使得所述第一转发隧道中转发的流量达到所述第一可用带宽触发流量分流或者所述第一转发隧道中的丢包率小于所述第一预定阈值。
对于所述第一预定带宽，可以是预设的固定带宽，也可以是根据丢包率确定的带宽，当所述第一预定带宽为根据丢包率确定的带宽时，可选的，本发明提供了一种如何计算第一可用带宽的方式，如所述公式1所示。
可见，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的额定带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所示第一转发隧道的额定带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性，由此提高了用户的使用体验。
当所述第一转发隧道由于丢包率较大，将所述第一转发隧道的额定带宽减少了所述第一预订带宽后，随着网络转发状态的改变以及丢包率的降低，所述第一网络设备需要将逐步恢复所述第一转发隧道的额定带宽，提高所述第一数据链路上的所述第一转发隧道转发流量的能力，以避免网络资源的浪费。本发明实施例通过对所述第一转发隧道丢包率大小的判断，对所述第一转发隧道的带宽进行增加。
在图7所示实施例的基础上，图9为本发明实施例提供的一种分流流量的装置结构图，如图9所示，所述分流流量装置700还包括带宽增加单元901，其中带宽增加单元901，用于当所述第一计算单元701的执行结果为所述第一丢包率小于第二预设阈值时，将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于或等于所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
首先需要注意的是，所述第一分流单元703和所述带宽增加单元901之间并没有限定触发的先后关系，所述带宽增加单元901并不需要一定要在触发所述第一分流单元703之后才能被触发，只要触发了所述带宽减少单元702，得到第一可用带宽后即可以触发所述带宽增加单元901，判断所示第一丢包率是否小于所述第二预设阈值。所述第二预设阈值小于所述第一预设阈值。
当所述第一丢包率小于所述第二预设阈值时，则代表当前所述第一转发链路的转发情况较好(比如说当前连接所述第一转发隧道的用户终端的数量减少等)，所述第一 网络设备可以将减少的手术室第一预设带宽的逐步的增加回来，在不考虑所述第一数据链路中从所述第一网络节点向第二网络节点中没有通过所述第一转发隧道转发的流量的情况下，增加后的第二可用带宽最大不能超过所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
当增加一次所述第二预订带宽后，如果所述第一丢包率持续小于所述第二预设阈值，可以持续对所述第一转发隧道的带宽进行增加，直到所述第一丢包率再一次超过所述第二预设阈值。
在所述第一数据链路中，一般不会只有从所述第一网络节点到所述第二网络节点的转发隧道，还可能有从所述第二网络节点到所述第一网络节点的转发隧道。
可选的，在图7的基础上，图10为本发明实施例提供的一种分流流量的装置结构图。所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分别是所述第二转发隧道的隧道端点。如图10所示，所述分流流量装置700，还包括第二计算单元1001和发送单元1002，其中第二计算单元1001，用于计算在所述预设周期内所述第二转发隧道的第二丢包率。发送单元1002，用于当所述第二计算单元的执行结果为所述第二丢包率大于第三预定阈值时，向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
需要注意的是，所述第二转发隧道是从所述第二网络节点向所述第一网络节点转发流量的，故当作为接收数据包的所述第一网络节点在计算预设周期内所述第二转发隧道的丢包率时，一般需要首先获取所述第二网络节点通过所述第二转发隧道发送的数据包总数，在下面的具体应用场景中，将对所述第二计算单元1001如何计算所述第二转发隧道的第二丢包率进行描述。
由于所述第一网络节点为所述第二转发隧道的目的端点，故当需要进行将所述第二转发隧道的额定带宽减少的操作时，所述第一网络节点需要指示所述第二网络节点，也就是所述第二转发隧道的源端点减少所述第二转发隧道的额定带宽，得到所述第三可用带宽。所述带宽通告报文可以是通用路由封装(Generic Routing Encapsulation，GRE)报文。
当所述第二转发隧道上转发的流量超出了所述第三可用带宽时，执行对超出部分的带宽进行分流操作的也是所述第二网络节点，也就是说，可选的，在如所示图10所示实施例的基础上，图11为本发明实施例提供的一种分流流量的装置结构图，所述分流流量装置1100包括：
第二分流单元1101，用于当所述第二转发隧道的转发流量超出所述第三可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
可选的，所述第一数据链路为具有数字用户线路DSL链路和网络协议IP链路。所述第二数据链路为长期演进LTE链路。
可选的，当所述第一网络节点为家庭网关HG时，所述第二网络节点为混合接入汇聚点HAAP。
当所述第一网络节点为HAAP时，所述第二网络节点为HG。
可见，第一网络节点在所述第一转发隧道的丢包率的较大的情况下减少所述第一转发隧道的额定带宽为第一可用带宽，使得当所述第一转发隧道转发的流量变大时，一旦超过所述第一可用带宽，就开始将超出的转发流量分流到第二数据链路上转发，由于所述第一可用带宽小于原本的所示第一转发隧道的额定带宽，转发流量会较早的超过所述第一可用带宽，提前了所述第一网络节点分流流量的分流时机，减少了在分流流量之前出现严重的丢包情况可能性，由此提高了用户的使用体验。当第一转发隧道上的丢包率逐渐减小到小于第二预设阈值时，则可以判定第一转发隧道的转发环境可以正常承载更多的流量转发。所述第一网络设备将已经减少后的第一转发隧道的第一可用带宽进一步增加第二预定带宽，以使得所述第一网络设备可以使用所述第一转发隧道比之前多转发第二预定带宽的流量，由此可以在所述第一转发隧道转发环境改善时，通过提高所述第一转发隧道的第一可用带宽，让所述第一转发隧道承载较多的流量转发，提高了网络资源的利用率。
实施例三
参阅图12，图12为本发明实施例提供的一种第一网络节点的硬件结构示意图，所述第一网络节点1200和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点1200向所述第二网络节点的第一转发隧道，所述第一网络节点1200和所述第二网络节点分别是所述第一转发隧道的隧道端点，所述第一网络节点1200包括存储器1201以及与所述存储器1201连接的处理器1202，所述存储器1201用于存储一组程序指令，所述处理器1202用于调用所述存储器1201存储的程序指令执行如下操作：
计算在预设周期内所述第一转发隧道的第一丢包率；
当所述第一丢包率大于第一预定阈值时，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；
当所述第一转发隧道的转发流量超出所述第一可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
可选地，所述处理器1202可以为中央处理器(Central Processing Unit，CPU)，所述存储器1201可以为随机存取存储器(Random Access Memory，RAM)类型的内部存储器。所述处理器1202和存储器1201可以集成为一个或多个独立的电路或硬件，如：专用集成电路(Application Specific Integrated Circuit，ASIC)。
本发明实施例中提到的第一网络节点、第一数据链路、第一转发隧道、第一丢包率、第一预定阈值、第一预定带宽、第一可用带宽和第一计数报文的“第一”只是用来做名字标识，并不代表顺序上的第一。该规则同样适用于“第二”和“第三”。
需要说明的是，本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
以上对本发明所提供的一种内存控制方法和装置进行了详细介绍，本文中应用了具体实施例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本发明的限制。

Claims (14)

  1. 一种分流流量的方法，其特征在于，第一网络节点和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点，该方法包括：
    所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率；
    当所述第一丢包率大于第一预定阈值时，所述第一网络节点将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；
    当所述第一转发隧道的转发流量超出所述第一可用带宽时，所述第一网络节点将超出的转发流量分流到所述第二数据链路上转发。
  2. 根据权利要求1所述的方法，其特征在于，还包括：
    当所述第一丢包率小于第二预设阈值时，所述第一网络节点将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于等于所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
  3. 根据权利要求1或2所述的方法，其特征在于，所述第一网络节点计算在预设周期内所述第一转发隧道的第一丢包率，具体包括：
    每隔所述预设周期，所述第一网络节点向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数；
    所述第一网络节点接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数；
    所述第一网络节点根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
  4. 根据权利要求1所述的方法，其特征在于，所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分别是所述第二转发隧道的隧道端点，该方法包括：
    所述第一网络节点计算在所述预设周期内所述第二转发隧道的第二丢包率；
    当所述第二丢包率大于第三预定阈值时，所述第一网络节点向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
  5. 根据权利要求4所述的方法，其特征在于，该方法还包括：
    当所述第二转发隧道的转发流量超出所述第三可用带宽时，所述第二网络节点将超出的转发流量分流到所述第二数据链路上转发。
  6. 根据权利要求1至5任意一项所述的方法，其特征在于，
    所述第一数据链路为具有数字用户线路DSL链路和网络协议IP链路；
    所述第二数据链路为长期演进LTE链路。
  7. 根据权利要求1至6任意一项所述的方法，其特征在于，
    当所述第一网络节点为家庭网关HG时，所述第二网络节点为混合接入汇聚点HAAP；
    当所述第一网络节点为HAAP时，所述第二网络节点为HG。
  8. 一种分流流量的装置，其特征在于，第一网络节点和第二网络节点之间连接有第一数据链路和第二数据链路，所述第一数据链路上建立有从所述第一网络节点向所述第二网络节点的第一转发隧道，所述第一网络节点和所述第二网络节点分别是所述第一转发隧道的隧道端点，包括：
    第一计算单元，用于计算在预设周期内所述第一转发隧道的第一丢包率；
    带宽减少单元，用于当所述第一计算单元的执行结果为所述第一丢包率大于第一预定阈值时，将所述第一转发隧道的额定带宽减少第一预定带宽，得到第一可用带宽；
    第一分流单元，用于当所述第一转发隧道的转发流量超出所述第一可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
  9. 根据权利要求8所述的装置，其特征在于，还包括：
    带宽增加单元，用于当所述第一计算单元的执行结果为所述第一丢包率小于第二预设阈值时，将所述第一可用带宽增加第二预定带宽，得到第二可用带宽，所述第二预定带宽小于等于所述第一预定带宽，所述第二可用带宽小于等于所述第一数据链路中从所述第一网络节点向第二网络节点的额定带宽。
  10. 根据权利要求8或9所述的装置，其特征在于，所述第一计算单元，包括：
    第一计数报文发送子单元，用于每隔所述预设周期，向所述第二网络节点发送第一计数报文，所述第一计数报文中携带所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数；
    第二计数报文接收子单元，用于接收所述第二网络节点返回的第二计数报文，所述第二计数报文中携带所述第一计数报文中携带的所述第一网络节点在所述预设周期内通过所述第一转发隧道向所述第二网络节点发送数据包的总数以及在所述预设周期内所述第二网络节点通过所述第一转发隧道接收所述第一网络节点发送的数据包的总数；
    第一丢包率计算子单元，用于根据所述第二计数报文中携带的所述第一网络节点向所述第二网络节点发送数据包的总数和所述第二网络节点接收所述第一网络节点发送的数据包的总数，计算对应所述预设周期的所述第一转发隧道的第一丢包率。
  11. 根据权利要求8所述的装置，其特征在于，所述第一数据链路中还建立有从所述第二网络节点向第一网络节点的第二转发隧道，所述第一网络节点和所述第二网络节点分别是所述第二转发隧道的隧道端点，该装置包括：
    第二计算单元，用于计算在所述预设周期内所述第二转发隧道的第二丢包率；
    发送单元，用于当所述第二计算单元的执行结果为所述第二丢包率大于第三预定阈值时，向所述第二网络节点发送带宽通告报文，所述带宽通告报文用于指示所述第二网络节点将所述第二转发隧道的额定带宽减少第三预定带宽，得到第三可用带宽。
  12. 根据权利要求11所述的装置，其特征在于，还包括：
    第二分流单元，用于当所述第二转发隧道的转发流量超出所述第三可用带宽时，将超出的转发流量分流到所述第二数据链路上转发。
  13. 根据权利要求8至12任意一项所述的装置，其特征在于，
    所述第一数据链路为具有数字用户线路DSL链路和网络协议IP链路；
    所述第二数据链路为长期演进LTE链路。
  14. 根据权利要求8至13任意一项所述的装置，其特征在于，
    当所述第一网络节点为家庭网关HG时，所述第二网络节点为混合接入汇聚点HAAP；
    当所述第一网络节点为HAAP时，所述第二网络节点为HG。
PCT/CN2015/085441 2014-08-05 2015-07-29 一种分流流量的方法和装置 Ceased WO2016019822A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP15829712.7A EP3166267B1 (en) 2014-08-05 2015-07-29 Method and apparatus for diverting flow
US15/423,922 US10251091B2 (en) 2014-08-05 2017-02-03 Traffic offloading method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410382683.1 2014-08-05
CN201410382683.1A CN104158761B (zh) 2014-08-05 2014-08-05 一种分流流量的方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/423,922 Continuation US10251091B2 (en) 2014-08-05 2017-02-03 Traffic offloading method and apparatus

Publications (1)

Publication Number Publication Date
WO2016019822A1 true WO2016019822A1 (zh) 2016-02-11

Family

ID=51884169

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/085441 Ceased WO2016019822A1 (zh) 2014-08-05 2015-07-29 一种分流流量的方法和装置

Country Status (4)

Country Link
US (1) US10251091B2 (zh)
EP (1) EP3166267B1 (zh)
CN (1) CN104158761B (zh)
WO (1) WO2016019822A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3550774A4 (en) * 2016-12-21 2019-12-04 Huawei Technologies Co., Ltd. PACKET TRANSMISSION METHOD AND HYBRID ACCESS GATEWAY

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9069727B2 (en) 2011-08-12 2015-06-30 Talari Networks Incorporated Adaptive private network with geographically redundant network control nodes
US10785117B2 (en) * 2009-06-11 2020-09-22 Talari Networks Incorporated Methods and apparatus for configuring a standby WAN link in an adaptive private network
CN104158761B (zh) 2014-08-05 2018-02-13 华为技术有限公司 一种分流流量的方法和装置
CN105743760B (zh) * 2014-12-12 2019-06-11 华为技术有限公司 一种流量切换方法和装置
CN106464567B (zh) * 2015-04-30 2020-04-28 华为技术有限公司 一种流量动态控制方法、设备及网关、融合接入汇聚点
CN106304397A (zh) * 2015-05-15 2017-01-04 北京华为数字技术有限公司 一种建立隧道的方法及相关设备
CN106982173A (zh) * 2016-01-18 2017-07-25 中兴通讯股份有限公司 一种隧道流量控制方法、装置和网络管理系统
CN105657748B (zh) * 2016-03-16 2020-06-26 华为技术有限公司 基于隧道绑定的通信方法和网络设备
CN106027405B (zh) * 2016-05-03 2020-04-10 浙江宇视科技有限公司 数据流的分流方法及装置
CN107454006B (zh) 2016-05-31 2020-07-24 华为技术有限公司 混合接入网络中的流量分发方法和装置
CN106789709B (zh) * 2016-12-21 2020-03-03 迈普通信技术股份有限公司 一种负载均衡的方法及装置
CN111740919B (zh) 2017-01-20 2023-08-22 华为技术有限公司 一种报负载分担方法及网络设备
CN108964933B (zh) * 2017-05-23 2020-09-29 华为技术有限公司 用于传输报文的方法、装置和系统
CN109218186B (zh) * 2017-07-05 2021-02-23 华为技术有限公司 一种多路径数据传输处理方法及网络设备
US12101652B2 (en) * 2018-01-18 2024-09-24 Cable Television Laboratories, Inc. Ad-hoc wireless mesh network system and methodology for failure reporting and emergency communications
US11025484B2 (en) 2018-01-18 2021-06-01 Cable Television Laboratories, Inc. Ad-hoc wireless mesh network system and methodology for failure reporting and emergency communications
CN110300431A (zh) * 2018-03-22 2019-10-01 华为技术有限公司 一种数据流量处理方法及相关网络设备
CN110351771A (zh) * 2018-04-04 2019-10-18 华为技术有限公司 提高网络吞吐率的方法、装置、系统、设备和介质
CN110505163B (zh) * 2018-05-18 2021-08-20 华为技术有限公司 传输报文的方法和装置
CN108616449A (zh) * 2018-08-14 2018-10-02 深圳市共进电子股份有限公司 数据传输方法、装置、设备和介质
CN109067607B (zh) * 2018-11-07 2020-09-18 网宿科技股份有限公司 一种调整额定带宽的方法和装置
CN109327283B (zh) * 2018-11-23 2021-06-22 深圳齐心好视通云计算有限公司 一种网络自适应的抗丢包方法、装置及终端设备
CN112448893A (zh) * 2019-08-29 2021-03-05 中兴通讯股份有限公司 一种链路带宽调节方法、装置及存储介质
CN111669340B (zh) * 2020-07-03 2021-06-22 全时云商务服务股份有限公司 发送带宽控制方法、装置、网络设备及可读存储介质
CN112019496B (zh) * 2020-07-06 2023-09-19 浙江华云信息科技有限公司 基于mqtt总线的主题安全订阅方法及装置
US11936566B2 (en) * 2020-12-18 2024-03-19 Dish Wireless L.L.C. Intelligent router bonding 5G telephony and digital subscriber line services

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101369972A (zh) * 2008-10-17 2009-02-18 华为技术有限公司 业务数据传输方法、系统及设备
CN101646196A (zh) * 2009-08-31 2010-02-10 华为技术有限公司 保证业务服务质量的方法及装置
US7680050B1 (en) * 2003-04-16 2010-03-16 Juniper Networks, Inc. Distributed admission control
CN102811465A (zh) * 2012-07-11 2012-12-05 北京邮电大学 异构网络下基于反馈信息的业务分流系统及方法
CN104158761A (zh) * 2014-08-05 2014-11-19 华为技术有限公司 一种分流流量的方法和装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1001574A1 (en) * 1998-11-10 2000-05-17 International Business Machines Corporation Method and system in a packet switching network for dynamically adjusting the bandwidth of a continuous bit rate virtual path connection according to the network load
US7787370B1 (en) * 2001-09-06 2010-08-31 Nortel Networks Limited Technique for adaptively load balancing connections in multi-link trunks
CN1842051A (zh) * 2005-03-30 2006-10-04 国际商业机器公司 流量均衡设备和方法以及使用它们的网络转发设备和方法
ES2582884T3 (es) * 2007-10-31 2016-09-15 Telefonaktiebolaget Lm Ericsson (Publ) Redes que tienen rutas múltiples entre nodos y nodos para tal red
US9787501B2 (en) * 2009-12-23 2017-10-10 Pismo Labs Technology Limited Methods and systems for transmitting packets through aggregated end-to-end connection
US8750098B2 (en) * 2010-07-28 2014-06-10 At&T Intellectual Property I, L.P. Femtocell service through a secondary connection
US9716659B2 (en) * 2011-03-23 2017-07-25 Hughes Network Systems, Llc System and method for providing improved quality of service over broadband networks
EP2728828A1 (en) * 2012-10-31 2014-05-07 British Telecommunications public limited company Session admission in a communications network
EP2728802B1 (en) * 2012-11-02 2020-08-12 Deutsche Telekom AG Method and system for network and service controlled hybrid access

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7680050B1 (en) * 2003-04-16 2010-03-16 Juniper Networks, Inc. Distributed admission control
CN101369972A (zh) * 2008-10-17 2009-02-18 华为技术有限公司 业务数据传输方法、系统及设备
CN101646196A (zh) * 2009-08-31 2010-02-10 华为技术有限公司 保证业务服务质量的方法及装置
CN102811465A (zh) * 2012-07-11 2012-12-05 北京邮电大学 异构网络下基于反馈信息的业务分流系统及方法
CN104158761A (zh) * 2014-08-05 2014-11-19 华为技术有限公司 一种分流流量的方法和装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3550774A4 (en) * 2016-12-21 2019-12-04 Huawei Technologies Co., Ltd. PACKET TRANSMISSION METHOD AND HYBRID ACCESS GATEWAY
US11025531B2 (en) 2016-12-21 2021-06-01 Huawei Technologies Co., Ltd. Packet transmission method and hybrid access gateway

Also Published As

Publication number Publication date
EP3166267A4 (en) 2017-08-16
US10251091B2 (en) 2019-04-02
CN104158761A (zh) 2014-11-19
EP3166267B1 (en) 2019-01-02
US20170150401A1 (en) 2017-05-25
CN104158761B (zh) 2018-02-13
EP3166267A1 (en) 2017-05-10

Similar Documents

Publication Publication Date Title
CN104158761B (zh) 一种分流流量的方法和装置
CN109309934B (zh) 一种拥塞控制方法及相关设备
JP6619815B2 (ja) アクセス制御装置、システム、及び方法
CN101447929B (zh) 一种流量选路方法、路由器和通信系统
CN102088413B (zh) 一种网络流量分流方法、网络节点及系统
CN105024940B (zh) 基于链路自适应的异构网络tcp拥塞控制方法
US12003407B2 (en) Resource usage in a multipath network
CN109617810B (zh) 数据传输方法及装置
CN108092888A (zh) 一种基于Overlay网络的传输方法、网关及传输系统
JP2018508151A (ja) 伝送制御プロトコルｔｃｐデータパケットを送信する方法及び装置、並びにシステム
US9912600B1 (en) Dynamic discovery of network packet size
WO2019179157A1 (zh) 一种数据流量处理方法及相关网络设备
US12010025B2 (en) System and method for accelerating or decelerating a data transport network protocol based on real time transport network congestion conditions
WO2018068516A1 (zh) 一种商业wifi用户动态带宽限制方法及系统
CN105915464B (zh) 一种快速简单的量化拥塞通告方法
US20220070736A1 (en) Traffic steering device
CN101754266A (zh) 调整传输速率、重定向路由的方法、系统及装置
WO2016090978A1 (zh) 一种流量切换方法和装置
CN104394093B (zh) 拥塞控制方法与无线网状网系统
CN103841043B (zh) 一种基于友好性的平行数据传输窗口机制
CN110505163B (zh) 传输报文的方法和装置
CN107612829B (zh) 一种数据报文的路径信息获取方法及设备
CN105991468B (zh) 一种Diameter拥塞响应的处理方法及装置
WO2018082518A1 (zh) 一种隧道调整方法和装置和计算机存储介质
CN116321281A (zh) 多接入数据分流方法、装置、电子设备及存储介质

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: 15829712

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2015829712

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2015829712

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE