WO2020259259A1 - 一种发送流量的方法和装置 - Google Patents

一种发送流量的方法和装置 Download PDF

Info

Publication number
WO2020259259A1
WO2020259259A1 PCT/CN2020/094792 CN2020094792W WO2020259259A1 WO 2020259259 A1 WO2020259259 A1 WO 2020259259A1 CN 2020094792 W CN2020094792 W CN 2020094792W WO 2020259259 A1 WO2020259259 A1 WO 2020259259A1
Authority
WO
WIPO (PCT)
Prior art keywords
network device
traffic
cost value
type
network
Prior art date
Application number
PCT/CN2020/094792
Other languages
English (en)
French (fr)
Inventor
王大为
刘京卿
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 EP20832399.8A priority Critical patent/EP3968585A4/en
Publication of WO2020259259A1 publication Critical patent/WO2020259259A1/zh
Priority to US17/562,668 priority patent/US20220124040A1/en

Links

Images

Classifications

    • 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
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/123—Evaluation of link metrics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/124—Shortest path evaluation using a combination of metrics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • 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
    • H04L47/122—Avoiding congestion; Recovering from congestion by diverting traffic away from congested entities
    • 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
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/245—Traffic characterised by specific attributes, e.g. priority or QoS using preemption
    • 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
    • H04L47/2458—Modification of priorities while in transit
    • 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
    • H04L47/28—Flow control; Congestion control in relation to timing considerations

Definitions

  • This application relates to the field of communication technology, and in particular to a method and device for sending traffic.
  • the shortest path first (English: shortest path first, referred to as SPF) algorithm is used to calculate the shortest path so that traffic is forwarded in the network according to the shortest path. Since all traffic occupies the shortest path in the network, once the traffic in the network increases, it is likely to cause congestion in the links between some network devices in the network.
  • the queue technology is adopted, that is, the traffic flowing into the first network device and flowing to the second network device is put into different priority queues, and the The traffic in the high-priority queue is forwarded to the second network device.
  • part of the traffic in the low-priority queue is forwarded to the second network device using the remaining bandwidth, and the traffic in the remaining low-priority queues throw away.
  • the above-mentioned queue technology is used to deal with link congestion between network devices, and part of the traffic needs to be discarded, and it cannot be guaranteed that all traffic entering the network will be forwarded in the network.
  • the embodiments of the present application provide a method and device for sending traffic, so as to avoid preempting bandwidth resources on congested links as much as possible by adjusting link overhead, so that part of the traffic reasonably utilizes other non-congested networks in the network.
  • the bandwidth resources on the link improve the traffic forwarding rate in the network.
  • an embodiment of the present application provides a method for sending traffic.
  • the method may include, for example, the first network device receives the first traffic, and determines that the first traffic belongs to the first traffic type according to the attribute information of the first traffic. ;
  • the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value, and the second network device is the first network
  • the device determines the next hop for sending the first traffic according to the first cost value;
  • the first network device determines, according to the adjusted second cost value, that the next hop for the first network device to send the first traffic is the third network device;
  • the first network device sends the first traffic to the third network device.
  • the link between the first network device and the second network device is congested, which may mean that the actual traffic rate from the first network device to the second network device is greater than that of the first network device and the second network device within the first preset time period.
  • the maximum physical link bandwidth between the second network devices; or, within the second preset time period, the average traffic rate from the first network device to the second network device is greater than the preset first physical link bandwidth; wherein , The preset first physical link bandwidth is less than the maximum physical link bandwidth.
  • the attribute information of the traffic is carried in the traffic and used to identify the traffic type to which the traffic belongs.
  • the attribute information may specifically be a traffic identifier corresponding to a priority identifier in the concept of Quality of Service (English: Quality of Service, abbreviated as: QoS), which is used to indicate the specific priority to which the traffic belongs in the QoS concept, for example : Flow identification used to indicate that the flow is BE flow; in another case, the attribute information can also be at least one flow feature in the five-tuple of the flow, for example: destination IP address, or source IP address and destination IP address.
  • QoS Quality of Service
  • the embodiment of the present application may further include: the first network device receives the first traffic and the second traffic , And determine according to the attribute information of the first traffic that the first traffic belongs to the first traffic type, and according to the attribute information of the second traffic, determine that the second traffic belongs to the second traffic type, and the first traffic type is different from the second traffic type;
  • the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value, which will correspond to the second traffic type
  • the third cost value is adjusted to the fourth cost value, and the second network device is the next hop determined by the first network device according to the first cost value and the third cost value to send the first traffic and the second traffic;
  • a network device determines according to the second cost value that the next hop to send the first traffic is the third network device, and according to the fourth cost value determines that the next hop to send the second traffic is the second
  • the first traffic avoids the congested link from the first network device to the second network device, and makes reasonable use of the network from the first network device to the third network device.
  • the non-congested link also ensures that the second traffic can be all forwarded on the original link from the first network device to the second network device, realizing effective forwarding of traffic in the network, and improving the traffic forwarding rate in the network.
  • the corresponding relationship between various traffic types and their corresponding cost values is pre-stored on the first network device.
  • a first mapping relationship is preset on the first network device, and the first mapping relationship includes the corresponding relationship between the first traffic type and the first cost value, and the relationship between the second traffic type and the third cost value
  • a second mapping relationship is preset on the first network device, and the second mapping relationship includes the corresponding relationship between the first traffic type, the first forwarding priority and the first cost value, and the second Correspondence between the traffic type, the second forwarding priority and the third cost value; where the first cost value is equal to the third cost value.
  • the flow type is used to distinguish different flows.
  • Traffic with different destination IP addresses corresponds to different traffic types. It should be noted that the traffic type can be flexibly configured in advance according to actual needs.
  • the above-mentioned first network device adjusting the first cost value corresponding to the first traffic type to the second cost value may specifically include: the first network device increases the first cost Value, the second cost value corresponding to the first traffic type is obtained, and the third network device and the second network device are different network devices; or, the first network device reduces the first cost value or keeps the first cost value unchanged , The second cost value corresponding to the first traffic type is obtained, and the third network device and the second network device are the same network device. It can be seen that different adjustment strategies can be flexibly adopted for the cost values corresponding to different traffic types through the pre-configured adjustment mode to realize effective forwarding of traffic.
  • the embodiment of the present application may further include: The correspondence between a traffic type and a second cost value is sent to other network devices.
  • each network device in the network can know the status of other network devices and links in the network, so that the network devices in the network can re-determine the forwarding path according to the current global situation, and provide data for the effective forwarding of traffic in the network. basis.
  • the embodiment of the present application may further include: the first network device according to The preset duration is used to determine whether the link from the first network device to the second network device is congested; when the link from the first network device to the second network device is not congested, the first network device will correspond to the first traffic type The second cost value of is restored to the first cost value.
  • the link from the first network device to the second network device is not congested, which may specifically be: within the third preset time period, the actual traffic rate from the first network device to the second network device is less than the preset second physical link Channel bandwidth; or, within the fourth preset time period, the average traffic rate from the first network device to the second network device is less than the preset third physical link bandwidth; wherein the preset third physical link bandwidth is less than the preset The first physical link bandwidth. Since the traffic in the network changes dynamically in real time, the time for link congestion to occur and maintain is not fixed.
  • the link is monitored for congestion, and when it is not congested, the link The cost value is restored to the cost value before adjustment, and the shortest path calculated by the cost value before adjustment is used to forward traffic, which can reduce the forwarding delay of traffic.
  • the first network device determines according to the second cost value that the next hop to send the first traffic is the third network device, which may specifically include: the first network device determines according to the second cost value The path for sending the first traffic; the first network device determines that the next hop of the first network device is the third network device based on the path for sending the first traffic. It can be seen that, according to the adjusted cost value, other paths different from the congested path are determined, to avoid discarding some packets, rationally utilize network resources, and improve the efficiency of packet forwarding.
  • an embodiment of the present application also provides a first network device for sending traffic, which may include: a first receiving unit, a first determining unit, a first adjusting unit, a second determining unit, and a first sending unit.
  • the first receiving unit is configured to receive the first traffic;
  • the first determining unit is configured to determine that the first traffic belongs to the first traffic type according to the attribute information of the first traffic;
  • the first adjusting unit is configured to be the first network device 2.
  • the link of the network device is congested, the first cost value corresponding to the first traffic type is adjusted to the second cost value, and the second network device is determined by the first network device to send the first traffic according to the first cost value.
  • Next hop a second determining unit, configured to determine, according to the second cost value, that the next hop for the first network device to send the first traffic is the third network device;
  • the first sending unit configured to send the first traffic to the third Internet equipment.
  • the device may further include: a second receiving unit, a third determining unit, a second adjusting unit, a fourth determining unit, and a second sending unit.
  • the second receiving unit is configured to receive the second flow; the third determining unit determines that the second flow belongs to the second flow type according to the attribute information of the second flow, and the first flow type is different from the second flow type; second The adjusting unit is configured to adjust the third cost value corresponding to the second traffic type to the fourth cost value when the link from the first network device to the second network device is congested, and the second network device is the first network
  • the device determines the next hop for sending the second traffic according to the third cost value; the fourth determining unit is configured to determine the next hop for sending the second traffic as the second network device according to the fourth cost value; the second sending unit uses To send the second traffic to the second network device.
  • the first mapping relationship is preset on the first network device, and the first mapping relationship includes the corresponding relationship between the first traffic type and the first cost value, and the relationship between the second traffic type and the third cost value.
  • a second mapping relationship is preset on the first network device, and the second mapping relationship includes the corresponding relationship between the first traffic type, the first forwarding priority and the first cost value, and the second traffic type, the first 2.
  • forwarding priority and third cost value where the first cost value is equal to the third cost value.
  • the first adjustment unit of the device is specifically configured to: increase the first cost value by the first network device to obtain the second cost value corresponding to the first traffic type, and 3.
  • the network device and the second network device are different network devices; or, the first network device reduces the first cost value or keeps the first cost value unchanged to obtain the second cost value corresponding to the first traffic type, and the third
  • the network device and the second network device are the same network device.
  • the device may further include: a third sending unit.
  • the third sending unit is configured to, after adjusting the first cost value corresponding to the first traffic type to the second cost value, send the correspondence between the first traffic type and the second cost value to other network devices .
  • the device may further include: a judgment unit and a third adjustment unit.
  • the judging unit is configured to judge whether the link from the first network device to the second network device is congested according to the preset time after adjusting the first cost value corresponding to the first traffic type to the second cost value;
  • the third adjustment unit is configured to restore the second cost value corresponding to the first traffic type to the first cost value when the link from the first network device to the second network device is not congested.
  • the second determining unit in the device includes a first determining subunit and a second determining subunit.
  • the first determining subunit is used to determine the path for sending the first traffic according to the second cost value; the second determining subunit is used to determine the next hop of the first network device based on the path of sending the first traffic.
  • the third network device includes a first determining subunit and a second determining subunit.
  • the device for sending traffic provided by the second aspect corresponds to the method for sending traffic provided by the first aspect. Therefore, the various possible implementation modes and technical effects achieved by the device for sending traffic provided by the second aspect , You can refer to the introduction of the method for sending traffic provided in the first aspect.
  • an embodiment of the present application also provides a network device, which includes a memory and a processor.
  • the memory is used to store program code;
  • the processor is used to run instructions in the program code, so that the network device executes the method for sending traffic in any one of the above-mentioned first aspects.
  • the embodiments of the present application also provide a computer program product, which when running on a computer, causes the computer to execute the method for sending traffic described in any one of the implementations of the first aspect.
  • the embodiments of the present application also provide a computer-readable storage medium that stores instructions in the computer-readable storage medium, which when run on a computer, causes the computer to execute any one of the aforementioned first aspects Implement the method of sending traffic described in the mode.
  • FIG. 1 is a schematic diagram of sending traffic in an application scenario in an embodiment of the application
  • Figure 2 is a schematic diagram of another sending traffic in an application scenario in an embodiment of the application
  • FIG. 3 is a schematic diagram of multiplexing router LSA under OSPFv2 protocol in an embodiment of this application;
  • FIG. 4 is a schematic diagram of extending router LSA under OSPFv3 protocol in an embodiment of this application.
  • FIG. 5 is a schematic flowchart of a method for sending traffic in an embodiment of this application.
  • FIG. 6 is a schematic flowchart of another method for sending traffic in an embodiment of this application.
  • FIG. 7 is a schematic flowchart of an example of a method for sending traffic in an embodiment of this application.
  • FIG. 8 is a schematic structural diagram of a device for sending traffic in an embodiment of the application.
  • Fig. 9 is a schematic structural diagram of a network device in an embodiment of the application.
  • the link bandwidth resources between network devices will face greater challenges. Specifically, when the link bandwidth allocated between network devices is smaller, the smaller link When the bandwidth cannot meet the larger traffic rate transmitted on the link, the link is congested. However, the existence of link congestion between network devices in the network usually leads to an increase in the transmission delay of the traffic, and even packet loss in severe cases, which causes the problem of degraded service quality or even unavailability in the network.
  • QoS Quality of Service
  • QoS Quality of Service
  • congestion management can be performed through queuing technology, that is, the traffic that will pass through the congested link is put into multiple queues, and each queue is pre-configured with different priorities.
  • the traffic in the queue with high priority takes priority to occupy the link bandwidth resources of the congested link, and is sent from the congested link; in the case of remaining bandwidth resources, according to the order of priority from high to low, partial low priority
  • the traffic in the level queue is sent from the congested link using the remaining link bandwidth resources.
  • the traffic in the remaining low priority queues is directly discarded.
  • shortest path first (English: shortest path first, referred to as: SPF) algorithm
  • SPF shortest path first algorithm
  • After calculating the shortest path according to the cost value cost on each link in the network it is determined that traffic 1 enters the network from port 1 of network device A (abbreviated as: port A1), and passes through port A2 ⁇ port B1 ⁇ port B5 ⁇ port C1 ⁇ C5 port ⁇ D1 port, forward the network from D3 port; and confirm that traffic 2 enters the network from the network device E1 port, passes E3 port ⁇ B2 port ⁇ B5 port ⁇ C1 port ⁇ C4 port ⁇ H2 port, and forwards from H3 port Out of the network.
  • the flow 1 is 70Gbps (the flow rate unit Gbps is 1000 megabits per second), including: guaranteed forwarding (English: assured forwarding, abbreviation: AF) traffic 10Gbps and best effort (English: best effort, abbreviation: BE) Flow 60Gbps; Flow 2 is 60Gbps flow, including: AF flow 30Gbps and BE flow 30Gbps; the maximum physical link bandwidth of link X is 70Gbps. In this way, flow 1 and flow 2 must pass through link X from port B5 to port C1, and the actual flow rate of flow 1 and flow 2 is greater than the maximum physical link bandwidth of link X, then link X is congested .
  • the specific process can be: the first step is to put traffic 1 and traffic 2 to be forwarded from link X into high priority queue 1 and low priority queue 1 and low priority respectively.
  • Level 2 queue specifically: put the 10Gbps AF traffic in traffic 1 and the 30Gbps AF traffic in traffic 2 into high priority queue 1, and put the 60Gbps BE traffic in traffic 1 and 30Gbps BE traffic in traffic 2 into low priority queue 1.
  • the second step is to send 30Gbps AF traffic in high priority queue 1 from link X, and select 30Gbps BE traffic from 90Gbps BE traffic in low priority queue 2 to occupy the remaining link X 30Gbps bandwidth resources; in the third step, because the bandwidth resources of the link X are all occupied, the remaining 60Gbps BE traffic in the low priority queue 2 is discarded to ensure the normal forwarding of high priority traffic.
  • queuing technology can manage link congestion to a certain extent, by locally discarding traffic in low-priority queues to ensure that high-priority traffic can be forwarded preferentially according to the shortest path, this method cannot Ensure that all traffic entering the network is forwarded in the network.
  • the network device by pre-saving the corresponding relationship between the traffic type and the cost value on the network device, after the network device receives the traffic, it can be based on the attribute information of the traffic (for example: the five-tuple information of the traffic Or the identification of the flow) to determine the flow type to which the flow belongs; when it is determined that the link from the network device to the next hop of the flow is congested, the network device can adjust the cost value corresponding to the flow type and base it on the The adjusted cost value re-determines the next hop for the network device to send the traffic.
  • the attribute information of the traffic for example: the five-tuple information of the traffic Or the identification of the flow
  • AF traffic is defined as traffic type 1
  • BE traffic is defined as traffic type 2.
  • two traffic types and corresponding cost values are stored on each network device.
  • network device B On network device B, a chain is stored in advance.
  • the mapping relationship of route X includes: the relationship between flow type 1 and cost value 10, and the relationship between flow type 2 and cost value 10.
  • traffic 1 and traffic 2 enter the network, based on the shortest path first algorithm or artificially configured paths, it can be determined that they pass through link X and the link X is congested. At this time, take traffic 1 as an example.
  • the network device B determines that there is 10Gbps of traffic a of traffic type 1 in the traffic 1, and there is 60Gbps of traffic b of traffic type 2; the network device B determines the overhead corresponding to the traffic type 2.
  • each network device can calculate the shortest path 1 corresponding to traffic type 1 and the corresponding to traffic type 2 based on the shortest path first algorithm Shortest path 2, and generate forwarding table 1 corresponding to traffic type 1 and forwarding table 2 corresponding to traffic type 2 on each network device; for network device B, forwarding table 1 on it indicates that the next hop is network device C, and forward Publish 2 indicates that the next hop is network device G. Then, network device B sends 10Gbps traffic a of traffic type 1 from port B2 to port C1 through link X, and sends 60Gbps traffic b of traffic type 2 from port B4 to port C1. Send to G2 port.
  • the corresponding relationship between the traffic type and the cost value of the related link of the network device is pre-stored.
  • the first network device stores: the relationship between traffic type 1 and cost value 1, traffic type 2 and cost value
  • the relationship between 2,..., the relationship between the traffic type N (N is an integer greater than 1) and the cost value n (n is a non-zero positive number) can be recorded as the mapping relationship 1 corresponding to link X;
  • the link Y from the first network device to the third network device is stored on the first network device: the relationship between traffic type 1 and cost value 1', the relationship between traffic type 2 and cost value 2',... ...
  • the relationship between the traffic type N (N is an integer greater than 1) and the cost value n'(n' is a non-zero positive number) can be recorded as the mapping relationship 2
  • the first network device before executing the method for sending traffic shown in FIG. 5 below, it may further include: S11, the first network device obtains the link sent from the first network device The corresponding relationship between the traffic type and the cost value; S12, the first network device saves the corresponding relationship between the traffic type and the cost value.
  • S11 may be specifically obtained according to configuration information configured on the first network device by a technician; it may also be obtained by receiving configuration information sent by other network devices and obtaining configuration information related to the first network device.
  • the corresponding relationship between the traffic type and the cost value may be stored on the network device in the form of a mapping relationship.
  • the mapping relationship saved on the network device can only include the correspondence between the traffic type and the cost value, or, in order to ensure the forwarding performance of certain traffic types, such as forwarding delay, it can also be configured for different traffic types.
  • the forwarding priority that is, the mapping relationship saved on the network device can also include the corresponding relationship between the traffic type, forwarding priority and cost value.
  • the traffic type with low forwarding priority can be corresponding Increase the cost value so that the traffic corresponding to the traffic type with low forwarding priority is forwarded from other links, and no longer occupy the bandwidth resources of the congested link, so as to ensure that the traffic of the traffic type with high forwarding priority is forwarded from the original path, reducing Its forwarding delay.
  • the corresponding relationship between the traffic type and the cost value will be stored as an example for description.
  • each network device in the network is global, that is, the standards for traffic classification in each network device are consistent, and the same traffic can be classified into the same traffic type. For example, if network device 1 receives traffic 1, it can be determined that the traffic belongs to traffic type 1. Then, if other network devices in the network receive traffic 1, it can also be determined that the traffic belongs to traffic type 1.
  • the correspondence between the traffic type and the cost value stored on each network device in the network can be global configuration information.
  • the corresponding relationship of all network devices in the network can be configured on one network device. Configure information, and publish the configuration information of all network devices in the network to inform each network device to obtain and save the corresponding relationship between the traffic type and the cost value.
  • the correspondence between the traffic type and the cost value can be carried in the Open Shortest Path First (English: Open Shortest Path First, OSPF) protocol extension or multiplexing link state broadcast service type (English: type of Service, abbreviation: TOS) field and distance metric field in the LSA.
  • Open Shortest Path First English: Open Shortest Path First, OSPF
  • TOS multiplexing link state broadcast service type
  • router link-state advertisement (English: router link-state advertisement, Abbreviation: TOS field and metric field in router LSA), in which #TOS field identifies the number of different traffic types or forwarding priorities; metric field identifies the link cost value; TOS identifies traffic type or forwarding priority; TOS metric identifies traffic The link cost value corresponding to the type or forwarding priority.
  • the correspondence between the traffic type and the cost value can also be carried in the link state routing intermediate system to intermediate system (English: intermediate system to intermediate system, abbreviated as: ISIS) protocol length type value ( English: Type-Length-Value, abbreviated as: TLV)
  • ISIS intermediate system to intermediate system
  • TLV Type-Length-Value
  • MT ID the reserved topology number
  • the two traffic types can be identified by the TLV reserved 3001 and 3002 respectively.
  • the content corresponding to the MT ID 3001 stores the cost value corresponding to a traffic type, and the corresponding MT ID 3002
  • the cost value corresponding to another traffic type is stored in the content.
  • the traffic type is used to distinguish different traffic. Specifically, it can be differentiated based on the priority identification in the QoS concept, such as AF traffic or BE traffic, or it can be differentiated based on traffic characteristics. Traffic, such as: traffic with different destination IP addresses corresponds to different traffic types. It should be noted that the traffic type can be flexibly configured in advance according to actual needs.
  • the traffic can be classified based on the priority in the QoS concept, and the priority identifier corresponding to each traffic type can be configured.
  • the traffic with priority from high to low in QoS can include: class selector (English: class selector, abbreviated as: CS) traffic, accelerated forwarding (English: Expedited Forwarding, abbreviated as: EF) traffic, AF traffic and BE traffic, If there are only two traffic types, then you can configure CS traffic to be associated with traffic type 1, and EF traffic, AF traffic, and BE traffic to be associated with traffic type 2.
  • CS traffic, EF traffic and traffic type 1 You can also configure CS traffic, EF traffic and traffic type 1 to associate, AF traffic , BE flow and flow type 2 are associated; CS flow, EF flow, AF flow and flow type 1 can also be configured, and BE flow is associated with flow type 2. If there are three types of traffic, then you can configure the association between CS traffic and traffic type 1, EF traffic and traffic type 2, AF traffic, BE traffic and traffic type 3; you can also configure CS traffic and traffic type 1, EF Flow, AF flow and flow type 2 are associated, BE flow is associated with flow type 3; CS flow, EF flow and flow type 1 can also be associated, AF flow is associated with flow type 2, and BE flow is associated with flow type 3.
  • the corresponding relationship between the traffic type and the identifier of the traffic can be preset on the network device. Then, after the network device receives the traffic, the traffic identifier can determine the traffic type to which the traffic belongs.
  • the identifier of the traffic may refer to the traffic that can uniquely identify which priority in QoS the traffic belongs to. It should be noted that the priority in the configuration process of the traffic type refers to the fixed priority defined for the traffic in QoS, which is different from the forwarding priority mentioned in the following embodiments of the application.
  • the flow can be allocated based on the flow characteristics, such as the five-tuple of the flow, and the corresponding flow type can be configured.
  • traffic with the same destination IP can be divided into one traffic type.
  • traffic with the same source and destination IP addresses can also be divided into one traffic type.
  • the corresponding relationship between the traffic type and the traffic characteristic can be preset on the network device. Then, after the network device receives the traffic, the traffic characteristic of the traffic can determine the traffic type to which the traffic belongs.
  • the link between the first network device and the second network device is congested. Specifically, the link meets at least one of the following two conditions: Within the set duration, the actual traffic rate from the first network device to the second network device is greater than the maximum physical link bandwidth between the first network device and the second network device; condition two, within the second preset duration , The average traffic rate from the first network device to the second network device is greater than the preset first physical link bandwidth; wherein the preset first physical link bandwidth is less than the maximum physical link bandwidth, for example: the preset first physical link bandwidth
  • the link bandwidth can be 80% of the maximum physical link bandwidth.
  • the first preset duration and the second preset duration may be equal or unequal, and can be flexibly set according to requirements.
  • the link between the first network device and the second network device is not congested.
  • the link may meet at least one of the following two conditions: Condition three, within the third preset time period, The actual traffic rate from the first network device to the second network device is less than the preset second physical link bandwidth; condition four, within the fourth preset time period, the average traffic rate from the first network device to the second network device is less than The preset third physical link bandwidth; wherein the preset third physical link bandwidth is less than the preset first physical link bandwidth.
  • the non-congested condition may specifically correspond to the congested condition, that is, the preset second physical link bandwidth is equal to the maximum physical link bandwidth, or, in order to improve the link non-congested condition, the preset second physical link may also be preset
  • the link bandwidth is set to 90% of the maximum physical link bandwidth, then the preset third physical link bandwidth may be 75% of the maximum physical link bandwidth.
  • condition one it is regarded as congested, and when condition three or four is met, it is regarded as not.
  • Congestion is another example: the link is deemed to be congested when condition two is satisfied, and it is deemed not congested when condition three or four is satisfied, which is not limited in the embodiment of this application.
  • Figure 5 is a flow diagram of a method for sending traffic in an embodiment of the application.
  • the method may include the following steps 501 to 505, for example:
  • Step 501 The first network device receives the first traffic, and determines that the first traffic belongs to the first traffic type according to the attribute information of the first traffic.
  • the first network device can obtain the traffic attribute of the first traffic, such as: the traffic identifier of the first traffic, or the quintuple of the first traffic; and then based on the traffic attribute Matching with the flow attributes corresponding to each preset flow type, if it matches, it can be determined that the first flow belongs to the flow type, and the successfully matched flow type can be determined as the first flow type to which the first flow belongs.
  • the traffic attribute of the first traffic such as: the traffic identifier of the first traffic, or the quintuple of the first traffic.
  • the attribute information of the flow is carried in the flow and used to identify the flow type to which the flow belongs.
  • the attribute information may specifically be a traffic identifier corresponding to the priority identifier in the QoS concept, which is used to indicate the specific priority to which the traffic belongs in the QoS concept, for example: a traffic identifier used to indicate that the traffic is BE traffic;
  • the attribute information may specifically also be at least one flow characteristic in the five-tuple of the flow, such as the destination IP address, or the source IP address and the destination IP address.
  • the traffic type in the correspondence between the traffic type and the cost value pre-stored on the first network device is a traffic identifier divided based on the priority identifier in the QoS concept, and the first network device is configured with CS traffic
  • EF traffic is associated with traffic type 1
  • AF traffic and BE traffic are associated with traffic type 2.
  • the first network device receives the first flow, it obtains the flow identifier of the first flow by parsing the first flow and indicates that the first flow is BE flow.
  • the BE flow can be separated from the flow type 1 and the flow type.
  • the flow identifier in 2 matches, and it is determined that the first flow and flow type 2 match successfully. Therefore, it can be determined that the first flow type to which the first flow belongs is flow type 2.
  • the first network device receives the first flow, it analyzes the first flow and obtains the flow characteristics of the first flow to indicate that the destination IP address of the first flow is 4.4.4.4, which can be achieved by comparing 4.4.4.4 with The flow characteristics in the flow type 1 and flow type 2 are matched, and it is determined that the first flow and the flow type 1 match successfully. Therefore, it can be determined that the first flow type to which the first flow belongs is the flow type 1.
  • the corresponding relationship between the traffic type and the cost value is stored on the first network device.
  • the corresponding relationship may be stored in the form of a first mapping relationship.
  • the corresponding relationship may also be stored in the form of a second mapping relationship.
  • Step 502 The first network device determines that the next hop for the first network device to send the first traffic is the second network device.
  • the first network device may first determine the first traffic type to which the first traffic belongs, and then determine the first cost value corresponding to the first traffic type; thus, based on the first cost value, pass the shortest
  • the path priority algorithm determines the shortest path of the first traffic in the network; viewing the shortest path, it can be determined that the next hop of the first network device when sending the first traffic is the second network device.
  • a preset path is configured for the first traffic in advance, and each network device that the preset path passes through is configured
  • preset path-related information which includes the configuration on the first network device: the next hop that sends the first traffic is the second network device.
  • the first network device After the first network device receives the first traffic, it may be determined that the first network device sends the first traffic based on the pre-configured information indicating "the next hop for sending the first traffic is the second network device". The next hop of one network device is the second network device.
  • Step 503 When the link between the first network device and the second network device is congested, the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value.
  • the first network device determines that the next hop that the received first traffic needs to pass is the second network device, and can directly transfer the first network device to the second network device.
  • a traffic is sent to the second network device; however, before the first network device sends the first traffic to the second network device, it is found that the link from the first network device to the second network device meets the preset link congestion condition, Then it can be determined that the link between the first network device and the second network device is congested.
  • the first network device can store the corresponding relationship between the traffic type and the cost value, and the first traffic type to which the first traffic belongs The corresponding first cost value is adjusted, and the adjusted first cost value is recorded as the second cost value, so that the forwarding path of the first traffic can be recalculated and determined based on the adjusted second cost value to avoid The current congested link.
  • the first network device adjusting the first cost value corresponding to the first traffic type to the second cost value specifically includes: the first network device increases the first cost value to obtain the value corresponding to the first traffic type
  • the second cost value for example, the first cost value 10 can be adjusted to 100, or the first cost value 10 can be increased by 9 times, and the second cost value obtained is 100. In this way, by increasing the first cost value to the second cost value, a data basis is provided for the subsequent first traffic to avoid the current congested link and forward from the non-congested link.
  • the first network device when the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value, the first network device can The adjusted correspondence between the first traffic type and the second cost value on a network device is sent to other network devices in the network.
  • the first network device may associate the second cost value with the identifier of the first traffic type and then publish it in the network.
  • the first network device may also associate the corresponding relationship between the first traffic type and the second cost value.
  • the first network device may also publish the entire mapping relationship stored on the first network device including the corresponding relationship between the first traffic type and the second cost value in the network.
  • Step 504 The first network device determines, according to the second cost value, that the next hop for the first network device to send the first traffic is the third network device.
  • Step 505 The first network device sends the first traffic to the third network device.
  • the first network device may determine the path to send the first traffic based on the second cost value based on the shortest path first algorithm; and then, based on the determined path to send the first traffic, determine that it is under the first network device.
  • One hop is the third network device; the first network device sends the first traffic to the third network device.
  • the first traffic can avoid the congested link from the first network device to the second network device, and the non-congested link from the first network device to the third network device in the network can be reasonably used to realize the first traffic in the network. Effective forwarding.
  • the embodiment of the present application may further include: when it is determined that the link between the first network device and the second network device is congested, and the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value After that, the first network device can monitor the congestion status of the link between the first network device and the second network device for a preset period of time. Once it is determined that the link from the first network device to the second network device is not congested, Then, the first network device restores the second cost value corresponding to the first traffic type to the first cost value.
  • the preset duration is a fixed constant, that is, the first network device periodically monitors the congestion status of the link from the first network device to the second network device, and the preset duration (for example: 10 minutes) is It is the preset period.
  • the first network device monitors the link from the first network device to the second network device for the third time The congestion status of the road, ..., and so on, until it is determined that the link from the first network device to the second network device is not congested, the first network device restores the second cost value corresponding to the first traffic type to The first cost value.
  • FIG. 6 shows a schematic flowchart of another method for sending traffic in an embodiment of the present application.
  • the method may specifically include the following steps 601 to 605:
  • Step 601 The first network device receives the first traffic and the second traffic, and determines that the first traffic belongs to the first traffic type according to the attribute information of the first traffic, and determines that the second traffic belongs to the second traffic according to the attribute information of the second traffic Types of.
  • the first network device can obtain the traffic attributes of the first traffic and the second traffic respectively; and then correspond to the preset traffic types based on the traffic attributes of the first traffic If it matches, it can be determined that the first traffic belongs to the traffic type.
  • the successfully matched traffic type can be determined as the first traffic type to which the first traffic belongs; similarly, based on the traffic attribute of the second traffic Matching with the traffic attributes corresponding to each preset traffic type, if it matches, it can be determined that the second traffic belongs to the traffic type, and the traffic type that is successfully matched can be determined as the second traffic type to which the second traffic belongs.
  • first flow type and the second flow type may be the same flow type or different flow types.
  • first flow type and the second flow type are two Two different types of traffic are explained.
  • the traffic type in the correspondence between the traffic type and the cost value pre-stored on the first network device is a traffic identifier divided based on the priority identifier in the QoS concept, and the first network device is configured with CS traffic
  • EF traffic is associated with traffic type 1
  • AF traffic and BE traffic are associated with traffic type 2.
  • the first network device receives the first flow and the second flow, it obtains the flow identifier of the first flow by analyzing the first flow and indicates that the first flow is BE flow.
  • the BE flow can be separated from the flow type. 1.
  • the flow identifier in the flow type 2 matches, and it is determined that the first flow and the flow type 2 match successfully.
  • the first flow type to which the first flow belongs is flow type 2; the second flow is analyzed to obtain the The flow identifier of the second flow indicates that the second flow is CS flow.
  • the CS flow can be matched with the flow identification in flow type 1 and flow type 2 to determine that the second flow and flow type 1 match successfully. Therefore, you can It is determined that the second flow type to which the second flow belongs is flow type 1.
  • the corresponding relationship between the traffic type and the cost value is stored on the first network device, which includes at least: the corresponding relationship between the first traffic type and the first cost value, and the corresponding relationship between the second traffic type and the third cost value , Where the first cost value is equal to the third cost value, or the third cost value may not be strictly equal to the first cost value, but the first cost value and the third cost value are required to be the shortest path determined by the shortest path first algorithm Both include the link between the first network device and the second network device.
  • Step 602 The first network device determines that the next hop of the first network device to send the first traffic and the second traffic is the second network device.
  • the first network device may determine that the received first traffic and second traffic are on the next hop network of the first network device based on the preset path or the shortest path calculated based on the first cost value and the third cost value.
  • the device is the second network device.
  • Step 603 When the link between the first network device and the second network device is congested, the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value; The third cost value corresponding to the second traffic type is adjusted to the fourth cost value.
  • the first network device determines that the next hop that the received first traffic and second traffic need to pass is the second network device, It is possible to directly send the first traffic and the second traffic to the second network device; however, before the first network device sends the first traffic to the second network device, it is found that the link from the first network device to the second network device satisfies
  • the preset link congestion condition can determine that the link between the first network device and the second network device is congested. At this time, the first network device can store the corresponding relationship between the traffic type and the cost value.
  • the first cost value corresponding to the first flow type to which a flow belongs is adjusted, and the adjusted first cost value is recorded as the second cost value.
  • the first network device can store the flow type and cost value on it.
  • the third cost value corresponding to the second flow type to which the second flow belongs is adjusted, and the adjusted third cost value is recorded as the fourth cost value, so that it can be based on the adjusted second cost value and
  • the fourth cost value is recalculated and the respective forwarding paths of the first flow and the second flow are determined to avoid the current congested link.
  • the first cost value corresponding to the first traffic type is adjusted to the second cost value; the first network device adjusts the third cost value corresponding to the second traffic type to the fourth cost value, where the adjustment , Can mean: increase, decrease or unchanged.
  • the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value, and adjusts the third cost value corresponding to the second traffic type to the fourth cost value, which may specifically include:
  • the first network device increases the first cost value, reduces or keeps the third cost value unchanged, that is, the second cost value is greater than the first cost value, and the fourth cost value is less than or equal to the third cost value.
  • two different paths corresponding to the first traffic and the second traffic can be calculated, where the path corresponding to the first traffic includes the link from the first network device to the third network device, and the second The path corresponding to the traffic includes a link from the first network device to the second network device.
  • This adjustment provides a data basis for the subsequent first traffic to avoid the current congested link, and the second traffic to continue to use the original link, and to rationally use the link resources in the network to forward traffic.
  • the first network device adjusts the first cost value corresponding to the first traffic type to the second cost value, and adjusts the third cost value corresponding to the second traffic type to the fourth cost value, specifically, Including: the first network device reduces or keeps the first cost value unchanged, and increases the third cost value, that is, the second cost value is less than or equal to the first cost value, and the fourth cost value is greater than the third cost value. Then, based on the shortest path first algorithm, two different paths corresponding to the first traffic and the second traffic can be calculated, where the path corresponding to the first traffic includes the link from the first network device to the second network device, and the second The path corresponding to the traffic includes the link from the first network device to the third network device. This adjustment provides a data basis for the subsequent second traffic to avoid the current congested link, the first traffic continues to use the original link, and the link resources in the network are reasonably used to forward traffic.
  • the first network device can adjust the first network device The subsequent correspondence between the first traffic type and the second cost value and the correspondence between the second traffic type and the fourth cost value are sent to other network devices in the network. It should be noted that, for the specific publishing method, refer to the description of the relevant part of the "Specific Method for the Network Device to Publish the Correspondence Between the Traffic Type and the Cost Value in the Network" before the introduction of this embodiment.
  • Step 604 The first network device determines, according to the second cost value, that the next hop that the first network device sends the first traffic is the third network device; according to the fourth cost value, it determines that the next hop that sends the second traffic is the The second network device.
  • Step 605 The first network device sends the first traffic to the third network device, and sends the second traffic to the second network device.
  • the first network device may determine the path for sending the first traffic based on the shortest path first algorithm according to the second cost value; and then, based on the determined path for sending the first traffic, determine that it is under the first network device.
  • One hop is the third network device.
  • the first network device may also determine the path for sending the second traffic based on the fourth cost value based on the shortest path first algorithm; and then, based on the determined path for sending the second traffic, determine that the next hop of the first network device is The second network device.
  • the first network device sends the first traffic and the second traffic to the third network device and the second network device respectively, so that the first traffic can avoid the congested link from the first network device to the second network device, Reasonably use the non-congested link from the first network device to the third network device in the network to realize the effective forwarding of the first traffic in the network.
  • the first network device may determine the path to send the first traffic based on the second cost value based on the shortest path first algorithm; and then determine the path to send the first traffic based on the determined path to send the first traffic The next hop is the second network device.
  • the first network device may also determine the path for sending the second traffic based on the fourth cost value based on the shortest path first algorithm; and then, based on the determined path for sending the second traffic, determine that the next hop of the first network device is The third network device.
  • the second traffic can avoid the congested link from the first network device to the second network device, and make rational use of the non-congested link from the first network device to the third network device in the network to achieve the first traffic in the network. Effective forwarding.
  • the method for sending two types of traffic is taken as an example for description, but the preset correspondence relationship in the embodiment of this application may specifically include two or Correspondence between more than 2 different traffic types and cost values.
  • the network device B in the network stores the mapping relationship corresponding to the link X from port B5 to port C1, which may specifically include: high forwarding priority-EF traffic type -Cost value 10, medium forwarding priority-AF traffic type-cost value 10 and low forwarding priority-BE traffic type-cost value 10; the corresponding cost values of the three traffic types of link Y from port B4 to port G2 20.
  • the cost value corresponding to the three traffic types of the link Z from the B3 port to the F3 port is also 20.
  • the maximum physical link bandwidth of link X is 50Gbps
  • the maximum physical link bandwidth of link Y is 70Gbps
  • the maximum physical link bandwidth of link Z is 40Gbps.
  • this example may specifically include:
  • step 701 the network device B determines that the flow 1, the flow 2, and the flow 3 pass through the link X at the same time, which will cause the link X to be congested.
  • step 702 the network device B adjusts the cost value of the traffic type corresponding to the middle and lower forwarding priority on the link X to 100, and publishes it on the network.
  • the network device B determines that the next hop of the network device B to send traffic 1 is still network device C according to the cost value of 10; and determines that the next hop of traffic 2 and traffic 3 is network device G according to the cost value of 100.
  • step 704 the network device B determines that the flow 2 and the flow 3 pass through the link Y at the same time, which will cause the link Y to be congested.
  • step 705 the network device B adjusts the cost value of the traffic type corresponding to the link Y with a lower forwarding priority to 100, and publishes it on the network.
  • mapping relationship corresponding to link Y stored on network device B may specifically include: high forwarding priority-EF traffic type-cost value 20, medium forwarding priority-AF traffic type-cost value 20, and low forwarding priority Class-BE traffic type-overhead value 100.
  • Step 706 The network device B determines that the next hop of the network device B to send traffic 2 is still network device G according to the cost value of 10; and determines that the next hop of traffic 3 is network device F according to the cost value of 100.
  • Step 707 Network device B sends traffic 1 to network device C via link X, sends traffic 2 to network device G via link Y, and sends traffic 3 to network device F via link Z.
  • the cost values corresponding to different traffic types are dynamically adjusted, so that the traffic corresponding to different traffic types uses different link bandwidth resources in the network, and no longer all preempt the same link. Therefore, the bandwidth resources on the non-congested links in the network can be reasonably used by the traffic to ensure that the traffic in the network can be effectively forwarded, and the traffic forwarding rate in the network is improved.
  • the classification of traffic types can be flexibly set according to actual needs; the storage format of the corresponding relationship saved on the network device and whether it includes forwarding priority can also be flexibly configured according to requirements;
  • the adjustment mechanism of the corresponding relationship saved on the device and the form of the adjustment on the network after adjustment can also be flexibly configured according to requirements; moreover, when more than two types of traffic are included, the adjustment mechanism when sending traffic It can also be flexibly configured according to the actual situation.
  • the embodiments of this application are only exemplary descriptions, as long as it can be realized that the embodiments of this application dynamically adjust the corresponding cost values of different traffic types, so that the traffic can reasonably use the bandwidth resources on the non-congested links in the network to ensure All traffic in the network can be effectively forwarded, and the effect of improving the traffic forwarding rate in the network falls within the protection scope of the embodiments of the present application.
  • FIG. 8 is a schematic structural diagram of a device for sending traffic according to an embodiment of the application.
  • the apparatus 800 may be a first network device, and the apparatus 800 may include: a first receiving unit 801, a first determining unit 802, a first adjusting unit 803, a second determining unit 804, and a first sending unit 805.
  • the first receiving unit 801 is configured to receive the first traffic.
  • the first determining unit 802 is configured to determine that the first traffic belongs to the first traffic type according to the attribute information of the first traffic.
  • the first adjustment unit 803 is configured to adjust the first cost value corresponding to the first traffic type to the second cost value when the link between the first network device and the second network device is congested, and the second network device is the first The network device determines the next hop for sending the first traffic according to the first cost value.
  • the second determining unit 804 is configured to determine, according to the second cost value, that the next hop for the first network device to send the first traffic is the third network device.
  • the first sending unit 805 is configured to send the first traffic to the third network device.
  • the device 800 is used to execute the steps corresponding to the embodiment shown in FIG. 5, so as to realize the sending of the traffic according to the method for sending the traffic provided in the embodiment of the present application.
  • the first receiving unit 801 and the first determining unit 802 correspond to the above step 501
  • the first adjusting unit 803 corresponds to the above step 502 to step 503
  • the second determining unit 804 corresponds to the above step 504
  • the first sending unit 805 can perform step 505 .
  • the device 800 may further include: a second receiving unit, a third determining unit, a second adjusting unit, a fourth determining unit, and a second sending unit.
  • the second receiving unit is used to receive the second flow.
  • the third determining unit determines, according to the attribute information of the second flow, that the second flow belongs to the second flow type, and the first flow type is different from the second flow type.
  • the second adjustment unit is configured to adjust the third cost value corresponding to the second traffic type to the fourth cost value when the link from the first network device to the second network device is congested, and the second network device is the first The network device determines the next hop for sending the second traffic according to the third cost value.
  • the fourth determining unit is configured to determine, according to the fourth cost value, that the next hop for sending the second traffic is the second network device.
  • the second sending unit is used to send the second traffic to the second network device.
  • the first mapping relationship is preset on the first network device, and the first mapping relationship includes the corresponding relationship between the first traffic type and the first cost value, and the relationship between the second traffic type and the third cost value.
  • a second mapping relationship is preset on the first network device, and the second mapping relationship includes the corresponding relationship between the first traffic type, the first forwarding priority and the first cost value, and the second traffic type, the first 2.
  • forwarding priority and third cost value where the first cost value is equal to the third cost value.
  • the first adjustment unit 803 of the apparatus 800 is specifically configured to: the first network device increases the first cost value to obtain the second cost value corresponding to the first traffic type, and the third The network device and the second network device are different network devices; or, the first network device reduces the first cost value or keeps the first cost value unchanged to obtain the second cost value corresponding to the first traffic type, and the third network The device and the second network device are the same network device.
  • the device 800 may further include: a third sending unit.
  • the third sending unit is configured to, after adjusting the first cost value corresponding to the first traffic type to the second cost value, send the correspondence between the first traffic type and the second cost value to other network devices .
  • the device 800 may further include: a judgment unit and a third adjustment unit.
  • the judging unit is configured to judge whether the link from the first network device to the second network device is congested according to the preset time after adjusting the first cost value corresponding to the first traffic type to the second cost value;
  • the third adjustment unit is configured to restore the second cost value corresponding to the first traffic type to the first cost value when the link from the first network device to the second network device is not congested.
  • the second determining unit 804 in the device 800 includes a first determining subunit and a second determining subunit.
  • the first determining subunit is used to determine the path for sending the first traffic according to the second cost value; the second determining subunit is used to determine the next hop of the first network device based on the path of sending the first traffic.
  • the third network device includes a first determining subunit and a second determining subunit.
  • the first determining subunit is used to determine the path for sending the first traffic according to the second cost value; the second determining subunit is used to determine the next hop of the first network device based on the path of sending the first traffic.
  • the device 800 for sending traffic provided by the embodiment of the present application corresponds to the above-mentioned method for sending traffic. Therefore, for various possible implementations and technical effects of the device 800, please refer to the aforementioned FIGS. 5 to 7 Corresponding to the relevant description of the embodiment.
  • the network device 900 includes a memory 901 and a processor 902.
  • the memory 901 is used for storing program codes
  • the processor 902 is used for running instructions in the program codes, so that the network device executes the method of sending traffic in any one of the implementations corresponding to the embodiments in FIG. 5 to FIG. 7.
  • the network device 900 may further include a communication interface, configured to receive the first traffic, and send the first traffic to the third network device.
  • the communication interface is used for data interaction with other network devices, for example: receiving first traffic from other network devices through the communication interface, or for example: sending second traffic to other network devices through the communication interface.
  • the processor 902 of the network device 900 is configured to run instructions in the program code stored on the memory 901, so that the network device executes the following method of sending traffic:
  • the first cost value corresponding to the first traffic type is adjusted to the second cost value, and the second network device is the first network device according to the first cost value.
  • the next hop of sending the first traffic determined by the value;
  • the embodiment of the present application also provides a computer program product, which when running on a computer, causes the computer to execute the method for sending traffic described in any one of the implementations corresponding to the embodiments in FIGS. 5 to 7.
  • the computer program product when it runs on a computer, such as a first network device, causes the computer to execute the following method of sending traffic:
  • the first cost value corresponding to the first traffic type is adjusted to the second cost value, and the second network device is the first network device according to the first cost value.
  • the next hop of sending the first traffic determined by the value;
  • the embodiments of the present application also provide a computer-readable storage medium that stores instructions in the computer-readable storage medium, and when it runs on a computer, the computer executes the corresponding embodiments in FIGS. 5-7. Any one of the methods for sending traffic described in the implementation mode.
  • the computer-readable storage medium has instructions stored in the computer-readable storage medium, and when it runs on a computer, such as a first network device, the computer executes the following method of sending traffic:
  • the first cost value corresponding to the first traffic type is adjusted to the second cost value, and the second network device is the first network device according to the first cost value.
  • the next hop of sending the first traffic determined by the value;
  • the computer software product can be stored in a storage medium, such as read-only memory (English: read-only memory, ROM)/RAM, magnetic disk, An optical disc, etc., includes a number of instructions to enable a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the method described in each embodiment of the application or some parts of the embodiment.
  • a computer device which may be a personal computer, a server, or a network communication device such as a router
  • the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.
  • the description is relatively simple, and for related parts, please refer to the partial description of the method embodiment.
  • the device embodiments described above are merely illustrative.
  • the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place. , Or it can be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the objectives of the solutions of the embodiments. Those of ordinary skill in the art can understand and implement it without creative work.

Landscapes

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

Abstract

本申请公开了一种发送流量的方法和装置，该方法包括：第一网络设备接收第一流量，并根据第一流量的属性信息确定第一流量属于第一流量类型；当第一网络设备到第二网络设备的链路出现拥塞时，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，该第二网络设备为第一网络设备根据第一开销值确定的发送该第一流量的下一跳；第一网络设备根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；第一网络设备将第一流量发送到第三网络设备。这样，通过调整拥塞链路上不同流量类型对应的开销值，使得部分类型的流量合理利用网络中的其他非拥塞链路上的带宽资源，确保网络中的流量都能被有效的转发，提高流量转发率。。

Description

一种发送流量的方法和装置
本申请要求于2019年6月27日提交中国专利局、申请号为201910570361.2、申请名称为“一种发送流量的方法和装置”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
本申请涉及通信技术领域，特别是涉及一种发送流量的方法和装置。
背景技术
随着网络的普及和互联网业务的发展，对网络的带宽资源需求不断提高。通常，在网络中基于网络设备之间的链路开销，利用最短路径优先(英文：shortest path first，简称：SPF)算法，计算最短路径，使得流量在该网络中按照该最短路径转发。由于所有的流量均抢占该网络中的最短路径，一旦网络中的流量增大，将很可能导致该网络中某些网络设备之间的链路出现拥塞。
目前，对于第一网络设备和第二网络设备之间的链路拥塞，采用队列技术，即，将流入第一网络设备且流向第二网络设备的流量，放入不同的优先级队列，优先将高优先级队列中的流量转发至第二网络设备，有剩余带宽资源的情况下，将部分低优先级队列中的流量利用剩余的带宽转发至第二网络设备，其余低优先级队列中的流量丢弃。
但是，采用上述队列技术处理网络设备之间出现链路拥塞的方式，需要丢弃部分流量，无法保证进入该网络中的全部流量在该网络中得到转发。
发明内容
基于此，本申请实施例提供了一种发送流量的方法和装置，以通过对链路开销的调整，尽量避免都抢占拥塞链路上的带宽资源，使得部分流量合理利用网络中的其他非拥塞链路上的带宽资源，提高网络中的流量转发率。
第一方面，本申请实施例提供了一种发送流量的方法，该方法例如可以包括：第一网络设备接收第一流量，并根据该第一流量的属性信息确定第一流量属于第一流量类型；当第一网络设备到第二网络设备的链路出现拥塞时，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，该第二网络设备为第一网络设备根据第一开销值确定的发送该第一流量的下一跳；第一网络设备根据该调整后的第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；第一网络设备将第一流量发送到第三网络设备。其中，第一网络设备到第二网络设备的链路发生拥塞，可以是指：在第一预设时长内，从第一网络设备到第二网络设备的实际流量速率均大于第一网络设备和所述第二网络设备之间的最大物理链路带宽；或者，在第二预设时长内，从第一网络设备到第二网络设备的平均流量速率大于预设第一物理链路带宽；其中，该预设第一物理链路带宽小于最大物理链路带宽。这样，在本申请实施例提供的方法中，通过调整拥塞链路上不同流量类型对应的开销值，尽量避免各种类型的流量都抢占拥塞链路上的带宽资源，使得部分类型的流量合理利 用网络中的其他非拥塞链路上的带宽资源，以确保网络中的流量都可以被有效的转发，提高网络中的流量转发率。
可以理解的是，流量的属性信息携带在流量中，用于标识流量所属的流量类型。一种情况下，该属性信息具体可以是服务质量(英文：Quality of Service，简称：QoS)概念中的优先级标识对应的流量标识，用于指示流量在QoS概念中属于的具体优先级，例如：用于指示流量为BE流量的流量标识；另一种情况下，该属性信息具体也可以是流量的五元组中至少一个流量特征，例如：目的IP地址，或，源IP地址和目的IP地址。
结合第一方面的一种具体的实现方式中，当第一网络设备接收到两个不同流量类型对应的流量时，本申请实施例还可以包括：第一网络设备接收第一流量和第二流量，并根据第一流量的属性信息确定该第一流量属于第一流量类型，根据第二流量的属性信息确定该第二流量属于第二流量类型，该第一流量类型和第二流量类型不同；当第一网络设备到第二网络设备经过的链路出现拥塞时，该第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，将与第二流量类型对应的第三开销值调整为第四开销值，该第二网络设备为第一网络设备根据第一开销值和第三开销值确定的发送第一流量和第二流量的下一跳；此时，第一网络设备根据第二开销值确定发送第一流量的下一跳为第三网络设备，根据第四开销值确定发送第二流量的下一跳为第二网络设备；那么，第一网络设备即可将第一流量发送给第三网络设备，将第二流量发送给第二网络设备。这样，通过调整拥塞链路上不同流量类型对应的开销值，使得第一流量避开第一网络设备到第二网络设备的拥塞链路，合理利用网络中第一网络设备到第三网络设备的非拥塞链路，而且，也确保了第二流量在第一网络设备到第二网络设备的原链路上可以全部被转发，实现流量在网络中的有效转发，提高网络中的流量转发率。
可以理解的是，第一网络设备上预先保存有各种流量类型和其对应的开销值的对应关系。例如：第一网络设备上预设第一映射关系，该第一映射关系包括所述第一流量类型和第一开销值之间的对应关系，以及第二流量类型和所述第三开销值之间的对应关系；或者，第一网络设备上预设第二映射关系，第二映射关系包括所述第一流量类型、第一转发优先级和第一开销值之间的对应关系，以及第二流量类型、第二转发优先级和第三开销值之间的对应关系；其中，第一开销值等于第三开销值。其中，流量类型，用于区分不同流量，具体可以是基于在QoS概念中的优先级标识区分得到的不同流量，如：AF流量或BE流量，也可以是基于流量特征区分得到的不同流量，如：目的IP地址不同的流量分别对应不同的流量类型。需要说明的是，流量类型可以根据实际需求预先被进行灵活的配置。
结合第一方面的另一种具体的实现方式，上述第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，具体可以包括：第一网络设备增大第一开销值，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为不同的网络设备；或者，第一网络设备减小第一开销值或保持第一开销值不变，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为相同的网络设备。可见，可以通过预先配置的调整方式灵活的对不同流量类型对应的开销值采取不同的调整策略，实现流量的有效转发。
结合第一方面的又一种实现方式，在第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值之后，本申请实施例还可以包括：第一网络设备将第一流量类型和第二开销值之间的对应关系，发送给其他网络设备。这样，网络中的各个网络设备均可以获 知网络中其他网络设备以及链路的情况，方便该网络中的网络设备根据当前的全局情况重新确定转发路径，为流量在网络中的有效转发提供了数据基础。
结合第一方面的再一种具体实现方式，在第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值之后，本申请实施例还可以包括：第一网络设备根据预设时长，判断从第一网络设备到第二网络设备的链路是否拥塞；当从第一网络设备到第二网络设备的链路不拥塞，则第一网络设备将与第一流量类型对应的第二开销值恢复为第一开销值。其中，第一网络设备到第二网络设备的链路不拥塞，具体可以是：在第三预设时长内，从第一网络设备到第二网络设备的实际流量速率小于预设第二物理链路带宽；或者，在第四预设时长内，从第一网络设备到第二网络设备的平均流量速率小于预设第三物理链路带宽；其中，预设第三物理链路带宽小于预设第一物理链路带宽。由于网络中的流量是实时动态变化的，所以，链路拥塞发生和维持的时间是不固定的，可见，本申请实施例中监控链路是否拥塞，并在不拥塞时，将该链路的开销值恢复到调整前的开销值，采用调整前的开销值计算的最短路径转发流量，可以降低流量的转发时延。
结合第一方面的又一种具体实现方式，第一网络设备根据第二开销值确定发送第一流量的下一跳为第三网络设备，具体可以包括：第一网络设备根据第二开销值确定发送第一流量的路径；第一网络设备基于发送第一流量的路径，确定在第一网络设备的下一跳为第三网络设备。可见，根据调整后的开销值确定不同于拥塞路径的其他路径，避免将部分报文丢弃，合理利用网络资源，提高了报文转发效率。
第二方面，本申请实施例还提供了一种发送流量的第一网络设备，可以包括：第一接收单元、第一确定单元、第一调整单元、第二确定单元和第一发送单元。第一接收单元，用于接收第一流量；第一确定单元，用于根据第一流量的属性信息确定第一流量属于第一流量类型；第一调整单元，用于当第一网络设备到第二网络设备的链路出现拥塞时，将与第一流量类型对应的第一开销值调整为第二开销值，第二网络设备为第一网络设备根据第一开销值确定的发送第一流量的下一跳；第二确定单元，用于根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；第一发送单元，用于将第一流量发送到第三网络设备。
结合第二方面的一种具体的实现方式，该装置还可以包括：第二接收单元、第三确定单元、第二调整单元、第四确定单元和第二发送单元。其中，第二接收单元，用于接收第二流量；第三确定单元，根据第二流量的属性信息确定第二流量属于第二流量类型，该第一流量类型和第二流量类型不同；第二调整单元，用于当第一网络设备到第二网络设备经过的链路出现拥塞时，将与第二流量类型对应的第三开销值调整为第四开销值，第二网络设备为第一网络设备根据第三开销值确定的发送第二流量的下一跳；第四确定单元，用于根据第四开销值确定发送第二流量的下一跳为第二网络设备；第二发送单元，用于将第二流量发送给第二网络设备。
可以理解的是，第一网络设备上预设第一映射关系，第一映射关系包括第一流量类型和第一开销值之间的对应关系，以及第二流量类型和第三开销值之间的对应关系；或者，第一网络设备上预设第二映射关系，第二映射关系包括第一流量类型、第一转发优先级和第一开销值之间的对应关系，以及第二流量类型、第二转发优先级和第三开销值之间的对应关系，其中，第一开销值等于第三开销值。
结合第二方面的另一种具体的实现方式，该装置的第一调整单元，具体用于：第一网络设备增大第一开销值，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为不同的网络设备；或者，第一网络设备减小第一开销值或保持第一开销值不变，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为相同的网络设备。
结合第二方面的又一种具体的实现方式，该装置还可以包括：第三发送单元。该第三发送单元，用于在将与第一流量类型对应的第一开销值调整为第二开销值之后，将第一流量类型和第二开销值之间的对应关系，发送给其他网络设备。
结合第二方面的一种具体的实现方式，该装置还可以包括：判断单元和第三调整单元。其中，判断单元，用于在将与第一流量类型对应的第一开销值调整为第二开销值之后，根据预设时长，判断从第一网络设备到第二网络设备的链路是否拥塞；第三调整单元，用于当从第一网络设备到第二网络设备的链路不拥塞，则，将与第一流量类型对应的第二开销值恢复为第一开销值。
结合第二方面的再一种具体的实现方式，该装置中的第二确定单元包括第一确定子单元和第二确定子单元。其中，第一确定子单元，用于根据第二开销值确定发送第一流量的路径；第二确定子单元，用于基于发送第一流量的路径，确定在第一网络设备的下一跳为第三网络设备。
需要说明的是，第二方面提供的发送流量的装置，对应于第一方面提供的发送流量的方法，故，第二方面提供的发送流量的装置的各种可能的实现方式以及达到的技术效果，可以参照前述第一方面提供的发送流量的方法的介绍。
第三方面，本申请实施例还提供了一种网络设备，该网络设备包括存储器和处理器。其中，存储器，用于存储程序代码；处理器，用于运行程序代码中的指令，使得网络设备执行以上第一方面中任意一种实现方式的发送流量的方法。
第四方面，本申请实施例还提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行前述第一方面中任意一种实现方式所述的发送流量的方法。
第五方面，本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得该计算机执行前述第一方面中任意一种实现方式所述的发送流量的方法。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请中记载的一些实施例，对于本领域普通技术人员来讲，还可以根据这些附图获得其他的附图。
图1为本申请实施例中一应用场景下发送流量的示意图；
图2为本申请实施例中一应用场景下另一发送流量的示意图；
图3为本申请实施例中一种OSPFv2协议下复用router LSA的示意图；
图4为本申请实施例中一种OSPFv3协议下扩展router LSA的示意图；
图5为本申请实施例中一种发送流量的方法的流程示意图；
图6为本申请实施例中另一种发送流量的方法的流程示意图；
图7为本申请实施例中发送流量的方法的一实例的流程示意图；
图8为本申请实施例中一种发送流量的装置的结构示意图；
图9为本申请实施例中一种网络设备的结构示意图。
具体实施方式
随着网络中传输数据量的增加，网络设备之间的链路带宽资源将面对更大的挑战，具体而言，当网络设备之间分配的链路带宽较小，而较小的链路带宽不能满足该链路上传输的较大的流量速率时，该链路出现拥塞。而网络中存在网络设备之间的链路拥塞，通常会导致流量的传输时延增大，严重时还会产生丢包，从而造成该网络中业务质量下降甚至不可用的问题。
服务质量(英文：Quality of Service，简称：QoS)作为网络的基本特性，一个重要的应用就是拥塞管理。通常，当网络设备之间的链路发生拥塞时，可以通过队列技术进行拥塞管理，即，将即将经过拥塞链路的流量按照放入多个队列中，各个队列预先配置有不同的优先级，将优先级高的队列中的流量优先占用该拥塞链路的链路带宽资源，从该拥塞链路发送；在有剩余带宽资源的情况下，按照优先级从高到低的顺序将部分低优先级队列中的流量利用剩余的链路带宽资源从该拥塞链路发送，在该拥塞链路无剩余的链路带宽资源时，将其余低优先级队列中的流量直接丢弃。
举例来说，对于图1所示的网络，包括网络设备A、B、C、D、D、E、F、G和H，基于最短路径优先(英文：shortest path first，简称：SPF)算法，根据网络中各段链路上的开销值cost计算最短路径后，确定流量1从网络设备A的1端口(简称：A1端口)进入该网络，经过A2端口→B1端口→B5端口→C1端口→C5端口→D1端口，从D3端口转发出该网络；并确定流量2从网络设备E1端口进入该网络，经过E3端口→B2端口→B5端口→C1端口→C4端口→H2端口，从H3端口转发出该网络。假设流量1为70Gbps(流量速率单位Gbps为1000兆比特每秒)的流量，包括：确保转发(英文：assured forwarding，简称：AF)流量10Gbps和尽力而为(英文：best effort，简称：BE)流量60Gbps；流量2为60Gbps的流量，包括：AF流量30Gbps和BE流量30Gbps；链路X的最大物理链路带宽为70Gbps。这样，流量1和流量2都要从B5端口经过链路X到达C1端口，且流量1和流量2的实际流量速率大于该链路X的最大物理链路带宽，那么，该链路X发生拥塞。
作为一个示例，若采用队列技术管理该链路X上的拥塞，具体过程可以是：第一步，将待从链路X转发的流量1和流量2分别放入高优先级队列1和低优先级队列2，具体为：将流量1中10Gbps的AF流量和流量2中30Gbps的AF流量放入高优先队列1，将流量1中60Gbps的BE流量和流量2中30Gbps的BE流量放入低优先队列2；第二步，将高优先级队列1中30Gbps的AF流量从链路X发送，并将低优先级队列2的90Gbps的BE流量中，选择30Gbps的BE流量占用该链路X剩余的30Gbps带宽资源；第三步，由于该链路X的带宽资源全部被占用，所以，将低优先级队列2的剩余60Gbps的BE流量丢弃，以保证高优先级流量的正常转发。
可见，采用队列技术虽然可以在一定程度上对链路拥塞进行管理，通过本地丢弃低优先级队列中的流量，确保高优先级的流量可以被优先的按照最短路径进行转发，但是，该方式无法保证进入该网络中的全部流量在该网络中得到转发。
基于此，在本申请实施例中，通过在网络设备上预先保存流量类型和开销值的对应关系，在网络设备接收到流量后，可以基于流量的属性信息(例如：该流量的五元组信息或者流量的标识)确定该流量属于的流量类型；在确定该流量从该网络设备到其下一跳的链路拥塞时，该网络设备可以将该流量类型对应的开销值进行调整，并基于该调整后的开销值重新确定该网络设备发送该流量的下一跳。这样，通过调整拥塞链路上不同流量类型对应的开销值，尽量避免各种类型的流量都抢占拥塞链路上的带宽资源，使得部分类型的流量合理利用网络中的其他非拥塞链路上的带宽资源，以确保网络中的流量都可以被有效的转发，提高网络中的流量转发率。
举例说明，仍然以图1对应的网络为场景，对本申请实施例提供的发送流量的方法进行说明。假设定义AF流量为流量类型1，BE流量为流量类型2，如图2所示，各网络设备上保存有两个流量类型和与之对应的开销值，在网络设备B上，预先保存有链路X的映射关系，该映射关系包括：流量类型1和开销值10的关系以及流量类型2和开销值10的关系。
作为一个示例，当流量1和流量2进入该网络时，基于最短路径优先算法或人为配置的路径，可以确定其经过链路X且该链路X发生拥塞，此时，以流量1为例，网络设备B根据该流量1中流量的标识，确定该流量1中有10Gbps的属于流量类型1的流量a，有60Gbps的属于流量类型2的流量b；该网络设备B将流量类型2对应的开销值调整为100，并将调整后的映射关系发布给网络中的其他网络设备；此时，各网络设备可以基于最短路径优先算法，计算出流量类型1对应的最短路径1和流量类型2对应的最短路径2，并在各网络设备上生成流量类型1对应的转发表1和流量类型2对应的转发表2；对于网络设备B，其上的转发表1指示下一跳为网络设备C，转发表2指示下一跳为网络设备G，那么，网络设备B将属于流量类型1的10Gbps流量a从B2端口经过链路X发送到C1端口，而将属于流量类型2的60Gbps流量b从B4端口发送到G2端口。这样，通过调整拥塞链路上不同流量类型对应的开销值，避免各种类型的流量都抢占拥塞链路上的带宽资源，使得部分类型的流量合理利用网络中的其他非拥塞链路上的带宽资源，确保了网络中的流量都可以被有效的转发，提高网络中的流量转发率。
可以理解的是，上述场景仅是本申请实施例提供的一个场景示例，本申请实施例并不限于此场景。
在介绍本申请实施例提供的发送流量的方法之前，先对网络设备上预设流量类型和开销值的过程，以及本申请实施例中涉及的相关概念进行相关说明。
可以理解的是，在网络中的各网络设备上，预先保存有该网络设备相关链路的流量类型和开销值之间的对应关系，例如：假设第一网络设备分别和第二网络设备、第三网络设备直连，那么，对于第一网络设备到第二网络设备的链路X，在第一网络设备上保存有：流量类型1和开销值1之间的关系、流量类型2和开销值2之间的关系、……、流量类型N(N为大于1的整数)和开销值n(n为非零正数)之间的关系，可以记作链路X对应的映射关系1；对于第一网络设备到第三网络设备的链路Y，在第一网络设备上保存有：流量类型1和开销值1’之间的关系、流量类型2和开销值2’之间的关系、……、流量类型N(N为大于1的整数)和开销值n’(n’为非零正数)之间的关系，可以记作链路Y对应的映射关系2。
为了确保在网络没有拥塞发生时，流量均可以基于最短路径转发，可以设置各流量类型对应的开销值相等，例如：可以设置开销值1＝开销值2＝……＝开销值n，开销值1’＝开销值2’＝……＝开销值n’。
具体实现时，对于网络中的第一网络设备，在执行下述图5所示的发送流量的方法之前，还可以包括：S11，第一网络设备获得从该第一网络设备发出的链路上的流量类型和开销值的对应关系；S12，第一网络设备保存该流量类型和开销值的对应关系。其中，S11具体可以是根据技术人员在该第一网络设备上进行配置的配置信息获得的；也可以是接收其他网络设备发送的配置信息，并从中获取与该第一网络设备相关的配置信息。
其中，网络设备上可以是以映射关系的形式保存流量类型和开销值的对应关系。网络设备上保存的映射关系可以只包括流量类型和开销值之间的对应关系，或者，为了确保某些流量类型的流量的转发性能，如：转发时延，也可以为不同的流量类型配置对应的转发优先级，即，网络设备上保存的映射关系也可以包括流量类型、转发优先级和开销值之间的对应关系，这样，当发生拥塞时，可以将转发优先级低的流量类型对应的开销值调大，使得转发优先级低的流量类型对应的流量从其他链路转发，不再占用该拥塞链路的带宽资源，从而确保转发优先级高的流量类型的流量从原路径转发，降低其转发时延。本申请实施例中将以保存流量类型和开销值的对应关系为例进行说明。
需要说明的是，网络中各网络设备上对于流量类型的定义和配置是全局性的，即，各个网络设备中对流量分类的标准一致，可以将相同的流量划分到相同的流量类型中。例如：若网络设备1接收到流量1，可以确定该流量属于流量类型1，那么，该网络中的其他网络设备接收到该流量1，同样可以确定该流量属于流量类型1。
需要说明的是，网络中各网络设备上保存的流量类型和开销值的对应关系，可以是全局性的配置信息，一种情况下，可以在一个网络设备上配置该网络中所有网络设备的该配置信息，并将该所有网络设备的该配置信息在该网络中发布，以告知各网络设备获取并保存其上相关的流量类型和开销值的对应关系。另一种情况下，也可以在各个网络设备上分别配置该网络设备相关的流量类型和开销值的对应关系，那么，需要各个网络设备将其上配置好的该网络设备相关的流量类型和开销值的对应关系在该网络中发布，以便根据全网的该配置信息进行全局性转发路径的确定。
对于网络设备在网络中发布流量类型和开销值之间的对应关系的具体方式，一种情况下，可以将流量类型和开销值之间的对应关系，携带在开放式最短路径优先(英文：Open Shortest Path First，简称：OSPF)协议的扩展或复用的链路状态广播LSA中的服务类型(英文：type of Service，简称：TOS)字段和距离metric字段中。例如，对于第二版开放式最短路径优先(英文：Open Shortest Path First version 2，简称：OSPFv2)协议，如图3所示，可以复用路由器链路状态广播(英文：router link-state advertisement，简称：router LSA)中的TOS字段和metric字段，其中，#TOS字段标识不同流量类型或转发优先级的数量；metric字段标识链路开销值；TOS标识流量类型或转发优先级；TOS metric标识流量类型或转发优先级对应的链路开销值，当包括流量类型1和流量类型2，且均对应的链路开销值分别为10和100时，那么，该router LSA中，Type字段之后分别为：#TOS＝2，metric＝10，TOS1＝1，TOS1 metric＝10，TOS2＝2；TOS2 metric＝100。又例如：对于第三版开放式最短路径优先(英文：Open Shortest Path First version 3，简称：OSPFv3)协议，如图4所示， 可以扩展路由器链路状态广播(英文：router link-state advertisement，简称：router LSA)，利用其中的TOS字段和metric字段，携带方式及其上面参见OSPFv2的相关说明。另一种情况下，还可以将流量类型和开销值之间的对应关系，携带在链路状态路由中间系统到中间系统(英文：intermediate system to intermediate system，简称：ISIS)协议的长度类型值(英文：Type-Length-Value，简称：TLV)预留的拓扑编号MT ID中。例如：假设预设2个流量类型，那么，可以通过TLV预留的3001和3002分别标识2个流量类型，该MT ID 3001对应的内容中保存一个流量类型对应的开销值，MT ID 3002对应的内容中保存另一个流量类型对应的开销值。
可以理解的是，流量类型，用于区分不同流量，具体可以是基于在QoS概念中的优先级标识区分得到的不同流量，如：AF流量或BE流量，也可以是基于流量特征区分得到的不同流量，如：目的IP地址不同的流量分别对应不同的流量类型。需要说明的是，流量类型可以根据实际需求预先被进行灵活的配置。
一种情况下，例如可以基于QoS概念里的优先级对流量进行分类，并配置各个流量类型对应的优先级标识。其中，QoS中优先级从高到低的流量可以包括：类选择码(英文：class selector，简称：CS)流量、加速转发(英文：Expedited Forwarding，简称：EF)流量、AF流量和BE流量，若只划分两个流量类型，那么，可以配置CS流量和流量类型1关联，EF流量、AF流量和BE流量和流量类型2关联；也可以配置CS流量、EF流量和流量类型1关联，AF流量、BE流量和流量类型2关联；还可以配置CS流量、EF流量、AF流量和流量类型1关联，BE流量和流量类型2关联。若划分三个流量类型，那么，可以配置CS流量和流量类型1关联，EF流量和流量类型2关联，AF流量、BE流量和流量类型3关联；也可以配置CS流量和流量类型1关联，EF流量、AF流量和流量类型2关联，BE流量和流量类型3关联；还可以配置CS流量、EF流量和流量类型1关联，AF流量和流量类型2关联，BE流量和流量类型3关联。若划分四个流量类型，那么，可以配置CS流量和流量类型1关联，EF流量和流量类型2关联，AF流量和流量类型3关联，BE流量和流量类型3关联。具体实现时，可以在网络设备上预设流量类型和流量的标识的对应关系，那么，当网络设备接收到流量后，通过该流量的标识即可确定该流量所属的流量类型。其中，流量的标识，可以是指可以唯一标识该流量属于QoS中哪个优先级对应的流量。需要说明的是，流量类型的配置过程中的该优先级，是指QoS中对流量定义的固定优先级，和下述本申请实施例中提及的转发优先级不同。
另一种情况下，例如还可以基于流量特征，如：流量的五元组，对流量进行分配，并配置对应的流量类型。如：可以将目的IP相同的流量划分到一个流量类型中，又如：还可以将源IP地址和目的IP地址均相同的流量划分到一个流量类型中。具体实现时，可以在网络设备上预设流量类型和流量特征的对应关系，那么，当网络设备接收到流量后，通过该流量的流量特征即可确定该流量所属的流量类型。
可以理解的是，在本申请实施例中，第一网络设备到第二网络设备的链路拥塞，具体可以是该链路满足下述两个条件中的至少一个：条件一、在第一预设时长内，从第一网络设备到第二网络设备的实际流量速率均大于第一网络设备和所述第二网络设备之间的最大物理链路带宽；条件二、在第二预设时长内，从第一网络设备到第二网络设备的平均流量速率大于预设第一物理链路带宽；其中，该预设第一物理链路带宽小于最大物理链路带 宽，例如：预设第一物理链路带宽可以是最大物理链路带宽的80％。其中，第一预设时长和第二预设时长，可以相等也可以不相等，可以根据需求进行灵活设置。
在本申请实施例中，第一网络设备到第二网络设备的链路不拥塞，具体可以是该链路满足下述两个条件中的至少一个：条件三、在第三预设时长内，从第一网络设备到第二网络设备的实际流量速率小于预设第二物理链路带宽；条件四、在第四预设时长内，从第一网络设备到第二网络设备的平均流量速率小于预设第三物理链路带宽；其中，预设第三物理链路带宽小于预设第一物理链路带宽。其中，该不拥塞的条件具体可以和拥塞条件对应，即，预设第二物理链路带宽等于最大物理链路带宽，或者，为了提高链路不拥塞的条件，也可以将预设第二物理链路带宽设置为最大物理链路带宽的90％，那么，该预设第三物理链路带宽可以为最大物理链路带宽的75％。
需要说明的是，链路拥塞和不拥塞的条件，可以根据实际情况和对网络的要求，进行灵活设置，如：链路在满足条件一即视作拥塞，满足条件三或四即视作不拥塞，又如：链路在满足条件二即视作拥塞，满足条件三或四即视作不拥塞，在本申请实施例中不作限定。
下面结合附图，通过实施例来详细说明本申请实施例中一种发送流量的方法的具体实现方式。
图5为本申请实施例中一种发送流量的方法的流量示意图。该方法例如可以包括下述步骤501～步骤505：
步骤501，第一网络设备接收第一流量，并根据第一流量的属性信息确定该第一流量属于第一流量类型。
具体实现时，第一网络设备接收到第一流量后，可以获得该第一流量的流量属性，如：该第一流量的流量标识，或该第一流量的五元组；再基于该流量属性与预设的各流量类型对应的流量属性进行匹配，若匹配，则可以确定该第一流量属于该流量类型，匹配成功的流量类型即可确定为第一流量所属的第一流量类型。
其中，流量的属性信息携带在流量中，用于标识流量所属的流量类型。一种情况下，该属性信息具体可以是QoS概念中的优先级标识对应的流量标识，用于指示流量在QoS概念中属于的具体优先级，例如：用于指示流量为BE流量的流量标识；另一种情况下，该属性信息具体也可以是流量的五元组中至少一个流量特征，例如：目的IP地址，或，源IP地址和目的IP地址。
例如：若第一网络设备上预先保存的流量类型和开销值的对应关系中的流量类型，是基于QoS概念中的优先级标识进行划分的流量标识，而且，第一网络设备中配置CS流量、EF流量和流量类型1关联，AF流量、BE流量和流量类型2关联。那么，当第一网络设备接收到第一流量后，通过解析第一流量，获得该第一流量的流量标识指示该第一流量为BE流量，可以通过将BE流量分别与流量类型1、流量类型2中的流量标识匹配，确定该第一流量和流量类型2匹配成功，故，可以确定该第一流量属于的第一流量类型为流量类型2。
又例如：若第一网络设备上预先保存的流量类型和开销值的对应关系中的流量类型，是基于流量特征中的目的IP地址确定的，而且，第一网络设备中至少配置有：4.4.4.4和流量类型1关联，8.8.8.8和流量类型2关联。那么，当第一网络设备接收到第一流量后，通过解析第一流量，获得该第一流量的流量特征指示该第一流量的目的IP地址为4.4.4.4， 可以通过将4.4.4.4分别与流量类型1、流量类型2中的流量特征匹配，确定该第一流量和流量类型1匹配成功，故，可以确定该第一流量属于的第一流量类型为流量类型1。
需要说明的是，本申请实施例中，在步骤501时，第一网络设备上保存有流量类型和开销值的对应关系，一种情况下，该对应关系可以是以第一映射关系的形式保存的。例如：该第一映射关系具体可以包括：第一流量类型和第一开销值之间的对应关系、第二流量类型和第三开销值之间的对应关系、……、第M流量类型和第M+1开销值之间的对应关系，其中，M为大于等于2的整数；初始状态下，可以设置各流量类型对应的开销值相等，即，第一开销值＝第三开销值＝……＝第M+1开销值。另一种情况下，该对应关系还可以是以第二映射关系的形式保存的。又例如：该第二映射关系包括第一流量类型、第一转发优先级和所述第一开销值之间的对应关系，所述第二流量类型、第二转发优先级和所述第三开销值之间的对应关系，……，第M流量类型、第M转发优先级和第M+1开销值之间的对应关系；初始状态下，可以设置各流量类型对应的开销值相等，即，第一开销值＝第三开销值＝……＝第M+1开销值，但转发优先级从高到低的关系为：第一转发优先级≥第二转发优先级≥……≥第M+1转发优先级。
步骤502，第一网络设备确定第一网络设备发送所述第一流量的下一跳为第二网络设备。
作为一个示例，第一网络设备接收到第一流量后，可以先确定第一流量所属的第一流量类型，再确定第一流量类型对应的第一开销值；从而基于第一开销值，通过最短路径优先算法，确定第一流量在该网络中的最短路径；查看该最短路径，可以确定发送该第一流量时第一网络设备的下一跳为第二网络设备。
作为另一个示例，第一流量在进入网络的同时或者之前，通过人工配置或者其他的路径选择算法，预先为该第一流量配置了预设路径，在该预设路径经过的每个网络设备配置有预设路径相关信息，其中包括在第一网络设备上配置的：发送第一流量的下一跳为第二网络设备。当第一网络设备接收到第一流量后，可以基于该第一网络设备上预先配置的指示“发送第一流量的下一跳为第二网络设备”的信息，确定发送该第一流量时第一网络设备的下一跳为第二网络设备。
步骤503，当第一网络设备到第二网络设备的链路出现拥塞时，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值。
具体实现时，当第一网络设备到第二网络设备的链路没有发现拥塞时，第一网络设备确定所接收的第一流量需要经过的下一跳为第二网络设备后，可以直接将第一流量发送至第二网络设备；但是，当第一网络设备发送第一流量至第二网络设备之前，发现第一网络设备到第二网络设备的链路满足了预设的链路拥塞条件，则可以确定第一网络设备到第二网络设备的链路出现拥塞，此时，第一网络设备可以对其上保存的流量类型和开销值的对应关系中，第一流量所属的第一流量类型对应的第一开销值进行调整，将调整后的第一开销值记作第二开销值，以使得可以基于调整后的第二开销值重新计算并确定该第一流量的转发路径，以避开当前的拥塞链路。
作为一个示例，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，具体包括：第一网络设备增大该第一开销值，得到与第一流量类型对应的第二开销值，例如：可以将第一开销值10调整为100，或，将第一开销值10增加其9倍，得到的第二开 销值均为100。如此，通过增大第一开销值为第二开销值，为后续第一流量避开当前的拥塞链路，从非拥塞链路转发提供了数据基础。
需要说明的是，由于流量类型和开销值的对应关系属于全局的配置信息，所以，当第一网络设备将第一流量类型对应的第一开销值调整为第二开销值后，可以将该第一网络设备上调整后的第一流量类型和第二开销值之间的对应关系，发送给该网络中的其他网络设备。具体实现时，第一网络设备可以将第二开销值和第一流量类型的标识进行关联后在网络中发布，第一网络设备也可以将第一流量类型和第二开销值之间的对应关系在网络中发布，或者，该第一网络设备也可以将包括第一流量类型和第二开销值之间对应关系的该第一网络设备上保存的映射关系整体在网络中发布。
需要说明的是，具体的发布方式可以参见介绍本实施例之前的“对于网络设备在网络中发布流量类型和开销值之间的对应关系的具体方式”相关部分的描述。
步骤504，第一网络设备根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备。
步骤505，第一网络设备将第一流量发送至第三网络设备。
具体实现时，第一网络设备可以根据第二开销值，基于最短路径优先算法，确定发送第一流量的路径；再基于所确定的该发送第一流量的路径，确定在第一网络设备的下一跳为第三网络设备；第一网络设备将第一流量发送至第三网络设备。如此，第一流量即可避开第一网络设备到第二网络设备的拥塞链路，合理利用网络中第一网络设备到第三网络设备的非拥塞链路，实现第一流量在网络中的有效转发。
这样，通过调整拥塞链路上不同流量类型对应的开销值，尽量避免各种类型的流量都抢占拥塞链路上的带宽资源，使得部分类型的流量合理利用网络中的其他非拥塞链路上的带宽资源，以确保网络中的流量都可以被有效的转发，提高网络中的流量转发率。
在一些具体的实现方式中，由于网络中的流量是实时动态变化的，所以，链路拥塞发生和维持的时间是不固定的，而且，若链路不拥塞，则采用调整前的开销值计算的最短路径转发流量，可以降低流量的转发时延。因此，本申请实施例还可以包括：在确定第一网络设备到第二网络设备的链路发生拥塞，且第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值之后，第一网络设备可以在预设时长，监测该第一网络设备到第二网络设备之间的链路的拥塞状况，一旦确定从第一网络设备到第二网络设备的链路不拥塞，则，第一网络设备将与第一流量类型对应的第二开销值恢复为第一开销值。
一种情况下，该预设时长为固定常数，即，第一网络设备周期性的监测从第一网络设备到第二网络设备的链路的拥塞状况，预设时长(例如：10分钟)即为预设周期。
另一种情况下，该预设时长为包括惩罚时长的变量，例如，在调整为第二开销值后的第5分钟(记作T1时刻)，第一网络设备第一次监测从第一网络设备到第二网络设备的链路的拥塞状况，若拥塞，则在T1时刻经过5*4＝20分钟(记作T2时刻)，第一网络设备第二次监测从第一网络设备到第二网络设备的链路的拥塞状况，若拥塞，则在T2时刻经过20*4＝80分钟(记作T3时刻)，第一网络设备第三次监测从第一网络设备到第二网络设备的链路的拥塞状况，……，以此类推，直到确定从第一网络设备到第二网络设备的链路不拥塞，则，第一网络设备将与第一流量类型对应的第二开销值恢复为第一开销值。
图6示出了本申请实施例中的另一种发送流量的方法的流程示意图。参见图6，该方法具体可以下述步骤601～步骤605：
步骤601，第一网络设备接收第一流量和第二流量，并根据第一流量的属性信息确定该第一流量属于第一流量类型，根据第二流量的属性信息确定第二流量属于第二流量类型。
具体实现时，第一网络设备接收到第一流量和第二流量后，可以分别获得该第一流量和第二流量的流量属性；再基于第一流量的流量属性与预设的各流量类型对应的流量属性进行匹配，若匹配，则可以确定该第一流量属于该流量类型，匹配成功的流量类型即可确定为第一流量所属的第一流量类型；同理，基于第二流量的流量属性与预设的各流量类型对应的流量属性进行匹配，若匹配，则可以确定该第二流量属于该流量类型，匹配成功的流量类型即可确定为第二流量所属的第二流量类型。
需要说明的是，第一流量类型和第二流量类型可以是相同的流量类型，也可以是不同的流量类型，本申请实施例的下述实现，以第一流量类型和第二流量类型为两个不同的流量类型的情况进行说明。
例如：若第一网络设备上预先保存的流量类型和开销值的对应关系中的流量类型，是基于QoS概念中的优先级标识进行划分的流量标识，而且，第一网络设备中配置CS流量、EF流量和流量类型1关联，AF流量、BE流量和流量类型2关联。那么，当第一网络设备接收到第一流量和第二流量后，通过解析第一流量，获得该第一流量的流量标识指示该第一流量为BE流量，可以通过将BE流量分别与流量类型1、流量类型2中的流量标识匹配，确定该第一流量和流量类型2匹配成功，故，可以确定该第一流量属于的第一流量类型为流量类型2；通过解析第二流量，获得该第二流量的流量标识指示该第二流量为CS流量，可以通过将CS流量分别与流量类型1、流量类型2中的流量标识匹配，确定该第二流量和流量类型1匹配成功，故，可以确定该第二流量属于的第二流量类型为流量类型1。
需要说明的是，第一网络设备上保存有流量类型和开销值的对应关系，至少包括：第一流量类型和第一开销值的对应关系、以及第二流量类型和第三开销值的对应关系，其中，第一开销值等于第三开销值，或者，第三开销值也可以和第一开销值不严格相等，但要求第一开销值和第三开销值利用最短路径优先算法确定的最短路径均包括第一网络设备到第二网络设备之间的链路。
步骤602，第一网络设备确定第一网络设备发送第一流量和第二流量的下一跳均为第二网络设备。
具体实现时，第一网络设备可以基于预设路径或者基于第一开销值和第三开销值计算的最短路径，确定所接收到第一流量和第二流量在第一网络设备的下一跳网络设备为第二网络设备。
步骤603，当第一网络设备到第二网络设备的链路出现拥塞时，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值；第一网络设备将与第二流量类型对应的第三开销值调整为第四开销值。
具体实现时，当第一网络设备到第二网络设备的链路没有发现拥塞时，第一网络设备确定所接收的第一流量和第二流量需要经过的下一跳为第二网络设备后，可以直接将第一流量和第二流量发送至第二网络设备；但是，当第一网络设备发送第一流量至第二网络设备之前，发现第一网络设备到第二网络设备的链路满足了预设的链路拥塞条件，则可以确 定第一网络设备到第二网络设备的链路出现拥塞，此时，第一网络设备可以对其上保存的流量类型和开销值的对应关系中，第一流量所属的第一流量类型对应的第一开销值进行调整，将调整后的第一开销值记作第二开销值，同理，第一网络设备可以对其上保存的流量类型和开销值的对应关系中，第二流量所属的第二流量类型对应的第三开销值进行调整，将调整后的第三开销值记作第四开销值，以使得可以基于调整后的第二开销值和第四开销值重新计算并确定该第一流量和第二流量各自的转发路径，以避开当前的拥塞链路。
需要说明的是，将与第一流量类型对应的第一开销值调整为第二开销值；第一网络设备将与第二流量类型对应的第三开销值调整为第四开销值，其中的调整，可以是指：增大、减小或不变。
作为一个示例，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，将与第二流量类型对应的第三开销值调整为第四开销值，具体可以包括：第一网络设备增大该第一开销值，减小或保持第三开销值不变，即，第二开销值大于第一开销值，第四开销值小于等于第三开销值。那么，基于最短路径优先算法，可以计算出第一流量和第二流量对应的两条不同的路径，其中，第一流量对应的路径包括第一网络设备到第三网络设备的链路，第二流量对应的路径包括第一网络设备到第二网络设备的链路。这样的调整，为后续第一流量避开当前的拥塞链路，第二流量继续利用原有链路，合理利用网络中的链路资源转发流量提供了数据基础。
作为另一个示例，第一网络设备将与第一流量类型对应的第一开销值调整为第二开销值，将与第二流量类型对应的第三开销值调整为第四开销值，具体还可以包括：第一网络设备减小或保持该第一开销值不变，增大第三开销值，即，第二开销值小于等于第一开销值，第四开销值大于第三开销值。那么，基于最短路径优先算法，可以计算出第一流量和第二流量对应的两条不同的路径，其中，第一流量对应的路径包括第一网络设备到第二网络设备的链路，第二流量对应的路径包括第一网络设备到第三网络设备的链路。这样的调整，为后续第二流量避开当前的拥塞链路，第一流量继续利用原有链路，合理利用网络中的链路资源转发流量提供了数据基础。
需要说明的是，由于流量类型和开销值的对应关系属于全局的配置信息，而且，流量类型也按照全局统一的流量划分标准获得的，所以，第一网络设备可以将该第一网络设备上调整后的第一流量类型和第二开销值之间的对应关系、以及第二流量类型和第四开销值之间的对应关系，发送给该网络中的其他网络设备。需要说明的是，具体的发布方式可以参见介绍本实施例之前的“对于网络设备在网络中发布流量类型和开销值之间的对应关系的具体方式”相关部分的描述。
步骤604，第一网络设备根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；根据第四开销值确定发送所述第二流量的下一跳为所述第二网络设备。
步骤605，第一网络设备将第一流量发送至第三网络设备，将第二流量发送至第二网络设备。
作为一个示例，第一网络设备可以根据第二开销值，基于最短路径优先算法，确定发送第一流量的路径；再基于所确定的该发送第一流量的路径，确定在第一网络设备的下一跳为第三网络设备。第一网络设备还可以根据第四开销值，基于最短路径优先算法，确定发送第二流量的路径；再基于所确定的该发送第二流量的路径，确定在第一网络设备的下 一跳为第二网络设备。这样，第一网络设备将第一流量和第二流量分别发送至第三网络设备和第二网络设备，如此，第一流量即可避开第一网络设备到第二网络设备的拥塞链路，合理利用网络中第一网络设备到第三网络设备的非拥塞链路，实现第一流量在网络中的有效转发。
作为另一个示例，第一网络设备可以根据第二开销值，基于最短路径优先算法，确定发送第一流量的路径；再基于所确定的该发送第一流量的路径，确定在第一网络设备的下一跳为第二网络设备。第一网络设备还可以根据第四开销值，基于最短路径优先算法，确定发送第二流量的路径；再基于所确定的该发送第二流量的路径，确定在第一网络设备的下一跳为第三网络设备。如此，第二流量即可避开第一网络设备到第二网络设备的拥塞链路，合理利用网络中第一网络设备到第三网络设备的非拥塞链路，实现第一流量在网络中的有效转发。
这样，通过调整拥塞链路上不同流量类型对应的开销值，尽量避免各种类型的流量都抢占拥塞链路上的带宽资源，使得部分类型的流量合理利用网络中的其他非拥塞链路上的带宽资源，以确保网络中的流量都可以被有效的转发，提高网络中的流量转发率。
可以理解的是，上述图5和图6所示的实施例中，以2个流量类型的流量的发送方法为例进行说明，但本申请实施例中预先设置的对应关系具体可以包括2个或2个以上不同的流量类型和开销值的对应关系。
为了使本申请实施例提供的发送流量的方法更加清楚，下面对一包括3个流量类型和开销值对应关系的具体实例进行说明。
以图1所示的网络架构为例介绍本申请实例，该网络中的网络设备B上保存B5端口到C1端口的链路X对应的映射关系，具体可以包括：高转发优先级-EF流量类型-开销值10、中转发优先级-AF流量类型-开销值10以及低转发优先级-BE流量类型-开销值10；B4端口到G2端口的链路Y的三个流量类型对应的开销值为20，B3端口到F3端口的链路Z的三个流量类型对应的开销值也为20。其中，链路X的最大物理链路带宽为50Gbps，链路Y的最大物理链路带宽为70Gbps，链路Z的最大物理链路带宽为40Gbps。
作为一个示例，若该网络中几乎同时接收流量速率10Gbps的EF流量1、流量速率60Gbps的AF流量2和流量速率30Gbps的BE流量3，根据最短路径优先算法，计算出该三个流量的路径均包括链路X，那么，当网络设备B接收到流量1、流量2和流量3时，如图7所示，本实例具体可以包括：
步骤701，网络设备B确定流量1、流量2和流量3同时经过链路X，将导致该链路X拥塞。
可以理解的是，由于流量1、流量2和流量3的流量速率之和等于10Gbps+60Gbps+30Gbps＝100Gbps，大于链路X的最大物理链路带宽50Gbps，故，可以确定链路X会出现拥塞。
步骤702，网络设备B将转发优先级中和转发优先级低在链路X上对应的流量类型的开销值调整为100，并在网络中发布。
步骤703，网络设备B根据开销值10确定该网络设备B发送流量1的下一跳仍然为网络设备C；根据开销值100确定发送流量2和流量3的下一跳为网络设备G。
步骤704，网络设备B确定流量2和流量3同时经过链路Y，将导致该链路Y拥塞。
可以理解的是，由于流量2和流量3的流量速率之和等于60Gbps+30Gbps＝90Gbps，大于链路Y的最大物理链路带宽70Gbps，故，可以确定链路Y会出现拥塞。
步骤705，网络设备B将转发优先级低在链路Y上对应的流量类型的开销值调整为100，并在网络中发布。
此时，网络设备B上保存的链路Y对应的映射关系，具体可以包括：高转发优先级-EF流量类型-开销值20、中转发优先级-AF流量类型-开销值20以及低转发优先级-BE流量类型-开销值100。
步骤706，网络设备B根据开销值10确定该网络设备B发送流量2的下一跳仍然为网络设备G；根据开销值100确定发送流量3的下一跳为网络设备F。
步骤707，网络设备B将流量1经过链路X发送至网络设备C，将流量2经过链路Y发送至网络设备G，将流量3经过链路Z发送至网络设备F。
这样，在网络设备之间链路发生拥塞时，通过动态调整不同流量类型对应的开销值，使得不同流量类型对应的流量利用该网络中不同的链路带宽资源，不再全部抢占同一条链路上的带宽资源，从而流量可以合理利用网络中非拥塞链路上的带宽资源，确保网络中的流量都可以被有效的转发，提高网络中的流量转发率。
需要说明的是，本申请实施例中，流量类型的划分可以根据实际需要进行灵活设置；网络设备上保存的对应关系的存储格式以及是否包括转发优先级，也可以根据需求灵活的进行配置；网络设备对其上保存的对应关系的调整机制，以及调整后在网络中发布的形式，也可以根据需求灵活的进行配置；而且，当包括两个以上的流量类型时，在发送流量时的调整机制也可以根据实际情况灵活配置。如上所示，本申请实施例仅是示例性的说明，只要可以实现本申请实施例通过动态调整不同流量类型对应的开销值，使得流量可以合理利用网络中非拥塞链路上的带宽资源，确保网络中的流量都可以被有效的转发，提高网络中的流量转发率的效果，均属于本申请实施例的保护范围。
图8为本申请实施例提供的一种发送流量的装置的结构示意图。该装置800可以是第一网络设备，该装置800可以包括：第一接收单元801、第一确定单元802、第一调整单元803、第二确定单元804和第一发送单元805。
其中，第一接收单元801，用于接收第一流量。
第一确定单元802用于根据第一流量的属性信息确定第一流量属于第一流量类型。
第一调整单元803用于当第一网络设备到第二网络设备的链路出现拥塞时，将与第一流量类型对应的第一开销值调整为第二开销值，第二网络设备为第一网络设备根据第一开销值确定的发送第一流量的下一跳。
第二确定单元804用于根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备。
第一发送单元805用于将第一流量发送到第三网络设备。
需要说明的是，该装置800用于执行上述图5所示实施例对应的各个步骤，以按照本申请实施例提供的流量发送方法实现流量的发送。其中，第一接收单元801和第一确定单元802对应上述步骤501，第一调整单元803对应上述步骤502～步骤503，第二确定单元804对应上述步骤504，第一发送单元805可以执行步骤505。
在一种具体的实现方式中，该装置800还可以包括：第二接收单元、第三确定单元、第二调整单元、第四确定单元和第二发送单元。
其中，第二接收单元用于接收第二流量。
第三确定单元，根据第二流量的属性信息确定第二流量属于第二流量类型，该第一流量类型和第二流量类型不同。
第二调整单元用于当第一网络设备到第二网络设备经过的链路出现拥塞时，将与第二流量类型对应的第三开销值调整为第四开销值，第二网络设备为第一网络设备根据第三开销值确定的发送第二流量的下一跳。
第四确定单元用于根据第四开销值确定发送第二流量的下一跳为第二网络设备。
第二发送单元用于将第二流量发送给第二网络设备。
可以理解的是，第一网络设备上预设第一映射关系，第一映射关系包括第一流量类型和第一开销值之间的对应关系，以及第二流量类型和第三开销值之间的对应关系；或者，第一网络设备上预设第二映射关系，第二映射关系包括第一流量类型、第一转发优先级和第一开销值之间的对应关系，以及第二流量类型、第二转发优先级和第三开销值之间的对应关系，其中，第一开销值等于第三开销值。
在另一种具体的实现方式中，该装置800的第一调整单元803，具体用于：第一网络设备增大第一开销值，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为不同的网络设备；或者，第一网络设备减小第一开销值或保持第一开销值不变，得到与第一流量类型对应的第二开销值，第三网络设备与第二网络设备为相同的网络设备。
在又一种具体的实现方式中，该装置800还可以包括：第三发送单元。该第三发送单元，用于在将与第一流量类型对应的第一开销值调整为第二开销值之后，将第一流量类型和第二开销值之间的对应关系，发送给其他网络设备。
在一种具体的实现方式中，该装置800还可以包括：判断单元和第三调整单元。其中，判断单元，用于在将与第一流量类型对应的第一开销值调整为第二开销值之后，根据预设时长，判断从第一网络设备到第二网络设备的链路是否拥塞；第三调整单元，用于当从第一网络设备到第二网络设备的链路不拥塞，则，将与第一流量类型对应的第二开销值恢复为第一开销值。
在再一种具体的实现方式，该装置800中的第二确定单元804包括第一确定子单元和第二确定子单元。其中，第一确定子单元，用于根据第二开销值确定发送第一流量的路径；第二确定子单元，用于基于发送第一流量的路径，确定在第一网络设备的下一跳为第三网络设备。
需要说明的是，本申请实施例提供的发送流量的装置800，对应于上述发送流量的方法，故，该装置800的各种可能的实现方式以及达到的技术效果，可以参照前述图5～图7对应实施例的相关描述。
此外，本申请实施例还提供了一种网络设备，如图9所示，该网络设备900包括存储器901和处理器902。其中，存储器901，用于存储程序代码；处理器902，用于运行程序代码中的指令，使得网络设备执行前述图5～图7对应实施例中任意一种实现方式的发送流量的方法。
示例的，该网络设备900还可以包括通信接口，用于接收第一流量，将所述第一流量发送到所述第三网络设备。通信接口用于和其他网络设备进行数据交互，例如：通过该通信接口从其他网络设备接收第一流量，又例如：通过该通信接口向其他网络设备发送第二流量。
示例的，该网络设备900的处理器902，用于运行存储器901上保存的程序代码中的指令，使得该网络设备执行下述发送流量的方法：
根据该第一流量的属性信息确定第一流量属于第一流量类型；
当第一网络设备到第二网络设备的链路出现拥塞时，将与第一流量类型对应的第一开销值调整为第二开销值，该第二网络设备为第一网络设备根据第一开销值确定的发送第一流量的下一跳；
根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；
此外，本申请实施例还提供了一种计算机程序产品，当其在计算机上运行时，使得计算机执行前述图5～图7对应实施例中任意一种实现方式所述的发送流量的方法。
示例的，该计算机程序产品，当其在计算机上运行时，例如第一网络设备上运行时，使得计算机执行下述发送流量的方法：
接收第一流量，并根据该第一流量的属性信息确定第一流量属于第一流量类型；
当第一网络设备到第二网络设备的链路出现拥塞时，将与第一流量类型对应的第一开销值调整为第二开销值，该第二网络设备为第一网络设备根据第一开销值确定的发送第一流量的下一跳；
根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；
将所述第一流量发送到所述第三网络设备。
此外，本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得该计算机执行前述图5～图7对应实施例中任意一种实现方式所述的发送流量的方法。
示例的，该计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，例如第一网络设备上运行时，使得该计算机执行下述发送流量的方法：
接收第一流量，并根据该第一流量的属性信息确定第一流量属于第一流量类型；
当第一网络设备到第二网络设备的链路出现拥塞时，将与第一流量类型对应的第一开销值调整为第二开销值，该第二网络设备为第一网络设备根据第一开销值确定的发送第一流量的下一跳；
根据第二开销值确定第一网络设备发送第一流量的下一跳为第三网络设备；
将所述第一流量发送到所述第三网络设备。
本申请实施例中提到的“第一开销值”、“第一网络设备”等名称中的“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”等。
通过以上的实施方式的描述可知，本领域的技术人员可以清楚地了解到上述实施例方法中的全部或部分步骤可借助软件加通用硬件平台的方式来实现。基于这样的理解，本申请的技术方案可以以软件产品的形式体现出来，该计算机软件产品可以存储在存储介质中，如只读存储器(英文：read-only memory，ROM)/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者诸如 路由器等网络通信设备)执行本申请各个实施例或者实施例的某些部分所述的方法。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于装置实施例而言，由于其基本相似于方法实施例，所以描述得比较简单，相关之处参见方法实施例的部分说明即可。以上所描述的装置实施例仅仅是示意性的，其中作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理模块，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模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下，即可以理解并实施。
以上所述仅是本申请示例性的实施方式，并非用于限定本申请的保护范围。
以上所述，仅为本发明的具体实施方式，但本发明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发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。

Claims (16)

  1. 一种发送流量的方法，其特征在于，包括：
    第一网络设备接收第一流量，并根据所述第一流量的属性信息确定所述第一流量属于第一流量类型；
    当所述第一网络设备到第二网络设备的链路出现拥塞时，所述第一网络设备将与所述第一流量类型对应的第一开销值调整为第二开销值，所述第二网络设备为所述第一网络设备根据所述第一开销值确定的发送所述第一流量的下一跳；
    所述第一网络设备根据所述第二开销值确定所述第一网络设备发送所述第一流量的下一跳为第三网络设备；
    所述第一网络设备将所述第一流量发送到所述第三网络设备。
  2. 根据权利要求1所述的方法，其特征在于，还包括：
    所述第一网络设备接收第二流量，并根据所述第二流量的属性信息确定所述第二流量属于第二流量类型，所述第一流量类型与所述第二流量类型不同；
    当所述第一网络设备到所述第二网络设备经过的链路出现拥塞时，所述第一网络设备将与所述第二流量类型对应的第三开销值调整为第四开销值，所述第二网络设备为所述第一网络设备根据所述第三开销值确定的发送所述第二流量的下一跳；
    所述第一网络设备根据所述第四开销值确定发送所述第二流量的下一跳为所述第二网络设备；
    所述第一网络设备将所述第二流量发送给所述第二网络设备。
  3. 根据权利要求1或2所述的方法，其特征在于，
    所述第一网络设备上预设第一映射关系，所述第一映射关系包括所述第一流量类型和所述第一开销值之间的对应关系，以及所述第二流量类型和所述第三开销值之间的对应关系；
    或者，所述第一网络设备上预设第二映射关系，所述第二映射关系包括所述第一流量类型、第一转发优先级和所述第一开销值之间的对应关系，以及所述第二流量类型、第二转发优先级和所述第三开销值之间的对应关系；
    其中，所述第一开销值等于所述第三开销值。
  4. 根据权利要求1～3任意一项所述的方法，其特征在于，所述第一网络设备将与所述第一流量类型对应的第一开销值调整为第二开销值，具体包括：
    所述第一网络设备增大所述第一开销值，得到与所述第一流量类型对应的所述第二开销值，所述第三网络设备与所述第二网络设备为不同的网络设备；
    或者，所述第一网络设备减小所述第一开销值或保持所述第一开销值不变，得到与所述第一流量类型对应的所述第二开销值，所述第三网络设备与所述第二网络设备为相同的网络设备。
  5. 根据权利要求1～4任意一项所述的方法，其特征在于，在所述第一网络设备将与所述第一流量类型对应的第一开销值调整为第二开销值之后，还包括：
    所述第一网络设备将所述第一流量类型和所述第二开销值之间的对应关系，发送给其他网络设备。
  6. 根据权利要求1～5任意一项所述的方法，其特征在于，在所述第一网络设备将与所述第一流量类型对应的第一开销值调整为第二开销值之后，还包括：
    所述第一网络设备根据预设时长，判断从所述第一网络设备到所述第二网络设备的链路是否拥塞；
    当从所述第一网络设备到所述第二网络设备的链路不拥塞，则所述第一网络设备将与所述第一流量类型对应的所述第二开销值恢复为所述第一开销值。
  7. 根据权利要求1～6任意一项所述的方法，其特征在于，所述第一网络设备根据所述第二开销值确定发送所述第一流量的下一跳为第三网络设备，包括：
    所述第一网络设备根据所述第二开销值确定发送所述第一流量的路径；
    所述第一网络设备基于所述发送所述第一流量的路径，确定所述第一网络设备的下一跳为所述第三网络设备。
  8. 一种发送流量的第一网络设备，其特征在于，包括：
    第一接收单元，用于接收第一流量；
    第一确定单元，用于根据所述第一流量的属性信息确定所述第一流量属于第一流量类型；
    第一调整单元，用于当所述第一网络设备到第二网络设备的链路出现拥塞时，将与所述第一流量类型对应的第一开销值调整为第二开销值，所述第二网络设备为所述第一网络设备根据所述第一开销值确定的发送所述第一流量的下一跳；
    第二确定单元，用于根据所述第二开销值确定所述第一网络设备发送所述第一流量的下一跳为第三网络设备；
    第一发送单元，用于将所述第一流量发送到所述第三网络设备。
  9. 根据权利要求8所述的装置，其特征在于，还包括：
    第二接收单元，用于接收第二流量；
    第三确定单元，用于根据所述第二流量的属性信息确定所述第二流量属于第二流量类型，所示第一流量类型与所述第二流量类型不同；
    第二调整单元，用于当所述第一网络设备到所述第二网络设备经过的链路出现拥塞时，将与所述第二流量类型对应的第三开销值调整为第四开销值，所述第二网络设备为所述第一网络设备根据所述第三开销值确定的发送所述第二流量的下一跳；
    第四确定单元，用于根据所述第四开销值确定发送所述第二流量的下一跳为所述第二网络设备；
    第二发送单元，用于将所述第二流量发送给所述第二网络设备。
  10. 根据权利要求8或9所述的装置，其特征在于，
    所述第一网络设备上预设第一映射关系，所述第一映射关系包括所述第一流量类型和所述第一开销值之间的对应关系，以及所述第二流量类型和所述第三开销值之间的对应关系；
    或者，所述第一网络设备上预设第二映射关系，所述第二映射关系包括所述第一流量类型、第一转发优先级和所述第一开销值之间的对应关系，以及所述第二流量类型、第二转发优先级和所述第三开销值之间的对应关系；
    其中，所述第一开销值等于所述第三开销值。
  11. 根据权利要求8～10任意一项所述的装置，其特征在于，所述第一调整单元，具体用于：
    所述第一网络设备增大所述第一开销值，得到与所述第一流量类型对应的所述第二开销值，所述第三网络设备与所述第二网络设备为不同的网络设备；
    或者，所述第一网络设备减小所述第一开销值或保持所述第一开销值不变，得到与所述第一流量类型对应的所述第二开销值，所述第三网络设备与所述第二网络设备为相同的网络设备。
  12. 根据权利要求8～11任意一项所述的装置，其特征在于，还包括：
    第三发送单元，用于在将与所述第一流量类型对应的第一开销值调整为第二开销值之后，将所述第一流量类型和所述第二开销值之间的对应关系，发送给其他网络设备。
  13. 根据权利要求8～12任意一项所述的装置，其特征在于，还包括：
    判断单元，用于在将与所述第一流量类型对应的第一开销值调整为第二开销值之后，根据预设时长，判断从所述第一网络设备到所述第二网络设备的链路是否拥塞；
    第三调整单元，用于当从所述第一网络设备到所述第二网络设备的链路不拥塞，则，将与所述第一流量类型对应的所述第二开销值恢复为所述第一开销值。
  14. 根据权利要求8～13任意一项所述的装置，其特征在于，所述第二确定单元包括：
    第一确定子单元，用于根据所述第二开销值确定发送所述第一流量的路径；
    第二确定子单元，用于基于所述发送所述第一流量的路径，确定在所述第一网络设备的下一跳为所述第三网络设备。
  15. 一种网络设备，其特征在于，所述网络设备包括存储器和处理器；
    所述存储器，用于存储程序代码；
    所述处理器，用于运行所述程序代码中的指令，使得所述网络设备执行以上权利要求1～7任一项所述的发送流量的方法。
  16. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得所述计算机执行以上权利要求1～7任一项所述的发送流量的方法。
PCT/CN2020/094792 2019-06-27 2020-06-07 一种发送流量的方法和装置 WO2020259259A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20832399.8A EP3968585A4 (en) 2019-06-27 2020-06-07 TRAFFIC TRANSMISSION METHOD AND DEVICE
US17/562,668 US20220124040A1 (en) 2019-06-27 2021-12-27 Traffic sending method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910570361.2A CN112152933B (zh) 2019-06-27 2019-06-27 一种发送流量的方法和装置
CN201910570361.2 2019-06-27

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/562,668 Continuation US20220124040A1 (en) 2019-06-27 2021-12-27 Traffic sending method and apparatus

Publications (1)

Publication Number Publication Date
WO2020259259A1 true WO2020259259A1 (zh) 2020-12-30

Family

ID=73868977

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/094792 WO2020259259A1 (zh) 2019-06-27 2020-06-07 一种发送流量的方法和装置

Country Status (4)

Country Link
US (1) US20220124040A1 (zh)
EP (1) EP3968585A4 (zh)
CN (1) CN112152933B (zh)
WO (1) WO2020259259A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113765796A (zh) * 2021-10-21 2021-12-07 迈普通信技术股份有限公司 流量转发控制方法及装置

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210243125A1 (en) * 2020-01-31 2021-08-05 Hewlett Packard Enterprise Development Lp System and method for avoiding congestion in a computer network
CN117081984B (zh) * 2023-09-27 2024-03-26 新华三技术有限公司 一种路由调整方法、装置及电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002082720A2 (en) * 2001-04-04 2002-10-17 Roke Manor Research Limited Automated network with tunnels and method of data routing
CN103166872A (zh) * 2013-03-07 2013-06-19 北京华为数字技术有限公司 网络中流量转发的方法和网络设备
CN103560971A (zh) * 2013-10-31 2014-02-05 华为技术有限公司 网络流量均衡控制方法和装置
CN107454015A (zh) * 2017-08-03 2017-12-08 山东省计算中心（国家超级计算济南中心） 一种基于OF‑DiffServ模型的QoS控制方法及系统
CN108063732A (zh) * 2016-11-07 2018-05-22 华为技术有限公司 报文传输方法、设备及系统
CN109617819A (zh) * 2019-01-29 2019-04-12 南京邮电大学 基于流量分类的软件定义回程网络路由方法

Family Cites Families (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2856050B2 (ja) * 1993-11-30 1999-02-10 日本電気株式会社 ルーティング制御方法
US6493317B1 (en) * 1998-12-18 2002-12-10 Cisco Technology, Inc. Traffic engineering technique for routing inter-class traffic in a computer network
US6831895B1 (en) * 1999-05-19 2004-12-14 Lucent Technologies Inc. Methods and devices for relieving congestion in hop-by-hop routed packet networks
US20040049595A1 (en) * 2001-12-04 2004-03-11 Mingzhou Sun System for proactive management of network routing
EP1335535A1 (en) * 2002-01-31 2003-08-13 BRITISH TELECOMMUNICATIONS public limited company Network service selection
WO2004068258A2 (de) * 2003-01-28 2004-08-12 Siemens Aktiengesellschaft Zuordnung von verteilungsgewichten zu verbindungsabschnitten in einem paketnetz mit verkehrsverteilung
US7409459B2 (en) * 2003-09-26 2008-08-05 Lucent Technologies Inc. System and method for provisioning QoS paths with restoration in a network
CN1756233B (zh) * 2004-09-30 2010-07-28 富士通株式会社 电信网络中的路由选择方法和装置
KR100717962B1 (ko) * 2005-07-15 2007-05-14 전자부품연구원 다수의 노드를 포함하는 무선 네트워크 시스템에서의데이터 전송 제어 방법 및 이를 이용한 센서 네트워크시스템 및 기록 매체
US20100142445A1 (en) * 2008-09-04 2010-06-10 Ludger Schlicht Environments for a mobile, broadband, routable internet
US8483194B1 (en) * 2009-01-21 2013-07-09 Aerohive Networks, Inc. Airtime-based scheduling
US8717893B2 (en) * 2010-11-29 2014-05-06 Verizon Patent And Licensing Inc. Network stabilizer
US8730817B2 (en) * 2010-12-07 2014-05-20 At&T Intellectual Property I, L.P. Methods and apparatus to determine network link weights
EP2730051B1 (en) * 2011-07-05 2015-06-17 Telefonaktiebolaget LM Ericsson (PUBL) Network congestion control with adaptive qos bit-rate differentiation
WO2016089921A1 (en) * 2014-12-01 2016-06-09 Luminus Networks Inc. System and method of discovering paths in a network
US20160234234A1 (en) * 2015-02-05 2016-08-11 Cisco Technology, Inc. Orchestrating the Use of Network Resources in Software Defined Networking Applications
EP3060000A1 (en) * 2015-02-20 2016-08-24 Thomson Licensing Adaptive load balancing in residential hybrid gateways
US10382341B2 (en) * 2015-03-20 2019-08-13 Juniper Networks, Inc. Label switched path preemption avoidance
GB2539976A (en) * 2015-06-30 2017-01-04 British Telecomm Communications network
WO2017092780A1 (en) * 2015-11-30 2017-06-08 Telecom Italia S.P.A. Dynamic configuration of routing paths in a data network
US10491501B2 (en) * 2016-02-08 2019-11-26 Ciena Corporation Traffic-adaptive network control systems and methods
CN105763467A (zh) * 2016-03-25 2016-07-13 杭州华三通信技术有限公司 流量切换方法及装置
US20190238449A1 (en) * 2017-01-31 2019-08-01 The Mode Group High performance software-defined core network
US10250658B2 (en) * 2017-03-17 2019-04-02 The Directv Group, Inc. Hybrid media stream delivery using multiple network connections
WO2018224157A1 (en) * 2017-06-08 2018-12-13 Telefonaktiebolaget Lm Ericsson (Publ) Optimal routing in a communications network
EP3487088B1 (en) * 2017-11-16 2023-03-01 Juniper Networks, Inc. Establishing communication with multiple networks to enable continuous communication coverage across the multiple networks
US10644978B2 (en) * 2017-11-22 2020-05-05 Hughes Network Systems, Llc Latency increase estimated rate limiter adjustment
US10742555B1 (en) * 2017-12-11 2020-08-11 Amazon Technologies, Inc. Network congestion detection and resolution
WO2019114959A1 (en) * 2017-12-13 2019-06-20 Telefonaktiebolaget Lm Ericsson (Publ) Methods in a telecommunications network
US10630579B1 (en) * 2018-01-16 2020-04-21 Amazon Technologies, Inc. Ensuring separate paths for network traffic between source devices and a destination device
CN108322925B (zh) * 2018-01-29 2021-02-19 东北大学 超密度异构融合网络中区分业务类型的传输路径计算方法
US11013056B2 (en) * 2018-04-13 2021-05-18 Qualcomm Incorporated Dynamic prioritization of uplink traffic
EP3815432B1 (en) * 2018-05-31 2024-03-06 Mobophiles, Inc., DBA Mobolize Systems and methods for dynamic channel bonding
US10827413B2 (en) * 2018-10-09 2020-11-03 At&T Intellectual Property I, L.P. Routing optimization based on historical network measures

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002082720A2 (en) * 2001-04-04 2002-10-17 Roke Manor Research Limited Automated network with tunnels and method of data routing
CN103166872A (zh) * 2013-03-07 2013-06-19 北京华为数字技术有限公司 网络中流量转发的方法和网络设备
CN103560971A (zh) * 2013-10-31 2014-02-05 华为技术有限公司 网络流量均衡控制方法和装置
CN108063732A (zh) * 2016-11-07 2018-05-22 华为技术有限公司 报文传输方法、设备及系统
CN107454015A (zh) * 2017-08-03 2017-12-08 山东省计算中心（国家超级计算济南中心） 一种基于OF‑DiffServ模型的QoS控制方法及系统
CN109617819A (zh) * 2019-01-29 2019-04-12 南京邮电大学 基于流量分类的软件定义回程网络路由方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3968585A4

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113765796A (zh) * 2021-10-21 2021-12-07 迈普通信技术股份有限公司 流量转发控制方法及装置
CN113765796B (zh) * 2021-10-21 2023-12-19 迈普通信技术股份有限公司 流量转发控制方法及装置

Also Published As

Publication number Publication date
US20220124040A1 (en) 2022-04-21
EP3968585A4 (en) 2022-07-13
EP3968585A1 (en) 2022-03-16
CN112152933A (zh) 2020-12-29
CN112152933B (zh) 2023-03-10

Similar Documents

Publication Publication Date Title
US10721156B2 (en) Technique for selecting a path computation element based on response time delay
CN112491713B (zh) 一种数据传输的控制方法及装置
US8369213B2 (en) Optimization of distributed tunnel rerouting in a computer network with path computation at an intermediate node
US7212490B1 (en) Dynamic load balancing for dual ring topology networks
WO2020259259A1 (zh) 一种发送流量的方法和装置
EP2713567B1 (en) Maintaining load balancing after service application with a network device
US7327675B1 (en) Fairness of capacity allocation for an MPLS-based VPN
US6963578B2 (en) Router
US8094555B2 (en) Dynamic weighted-fair load-balancing
US7948883B1 (en) Applying router quality of service on a cable modem interface on a per-service-flow basis
US8717893B2 (en) Network stabilizer
US20050007954A1 (en) Network device and method for categorizing packet data flows and loading balancing for packet data flows
US7551569B2 (en) Efficient tunnel placement in a computer network using distributed synchronization
US20050243723A1 (en) Multi-parameter load balancing device for a label switched communications network peripheral device
JP2008131240A (ja) ネットワークシステム、その装置及び方法
US7742416B2 (en) Control of preemption-based beat-down effect
JP2013034164A (ja) 中継装置、中継方法
EP3884616B1 (en) Segment routing network
US8264957B1 (en) Control of preemption-based beat-down effect
JP2006197473A (ja) ノード
KR20130085541A (ko) 패킷 전송 장치 및 패킷 전송 방법
JP4342395B2 (ja) パケット中継方法及び装置
Nakayama Rate-based path selection for shortest path bridging in access networks
Bai et al. Dynamic end-to-end QoS support for video over the Internet
KR20050001972A (ko) 제한 기반 라우팅 다중 프로토콜 레이블 스위칭 망에서최선형 서비스를 위한 라우팅 방법

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: 20832399

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020832399

Country of ref document: EP

Effective date: 20211209

NENP Non-entry into the national phase

Ref country code: DE