WO2022111325A1 - 通信方法、装置及系统 - Google Patents

通信方法、装置及系统 Download PDF

Info

Publication number
WO2022111325A1
WO2022111325A1 PCT/CN2021/130687 CN2021130687W WO2022111325A1 WO 2022111325 A1 WO2022111325 A1 WO 2022111325A1 CN 2021130687 W CN2021130687 W CN 2021130687W WO 2022111325 A1 WO2022111325 A1 WO 2022111325A1
Authority
WO
WIPO (PCT)
Prior art keywords
speed limit
limit point
bandwidth
physical link
branch
Prior art date
Application number
PCT/CN2021/130687
Other languages
English (en)
French (fr)
Inventor
刘云
汪大勇
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2022111325A1 publication Critical patent/WO2022111325A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/24—Multipath
    • H04L45/245—Link aggregation, e.g. trunking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/24—Multipath
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/22—Traffic shaping
    • 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
    • H04L47/22—Traffic shaping
    • H04L47/225—Determination of shaping rate, e.g. using a moving window
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

  • the present application relates to the field of communication, and in particular, to a communication method, device and system.
  • PLA Physical Link Aggregation
  • a communication system based on the PLA technology includes an aggregated link, and a sender network element (Network Element, NE) and a receiver network element connected through the aggregated link.
  • the aggregated link includes one or more member links configured on each of the multiple physical links.
  • the network element at the sending end is used for slicing service packets to obtain multiple sliced packets, and then distributes the multiple sliced packets to multiple member links in the aggregated link.
  • the network element at the receiving end is used for receiving and sorting multiple sliced packets, and then recombining the sorted sliced packets to obtain service packets. If any sliced packet is discarded during transmission, the reassembly of the sliced packet will fail, resulting in random packet loss.
  • the network element at the sending end passes a speed limit (also called traffic shaping) to slice through the member link.
  • the bandwidth occupied by packets is set to be less than or equal to the bandwidth of the member link.
  • the bandwidth of the member link is the bandwidth set by the staff for speed limit.
  • the staff needs to calculate the speed limit bandwidth of each speed limit point on the member link, and determine the minimum value of the speed limit bandwidth of each speed limit point as the bandwidth of the member link.
  • the rate-limiting bandwidth of each rate-limiting point needs to be comprehensively determined based on the bandwidth of each rate-limiting point, overhead bits, the minimum length of the sliced packets passing through the rate-limiting point, and the bandwidth of adjacent rate-limiting points, etc.
  • the labor cost is high. . As a result, the process of obtaining the bandwidth used for speed limiting is complicated, and the obtaining efficiency is low.
  • Embodiments of the present application provide a communication method, device, and system to improve rate limiting efficiency.
  • a communication method is provided, which is applied to a network element at a sending end, and the method includes: for each physical link in a plurality of physical links bound by PLA, the network element at the sending end obtains the multiple physical links on the physical link.
  • Speed limit point information for each speed limit point. Wherein, the speed limit point information includes bandwidth.
  • the network element at the sending end limits the rate of the sliced packets to be transmitted on the physical link based on the bandwidth of multiple rate limiting points on the physical link.
  • the network element at the sending end uses the obtained bandwidths of multiple speed-limiting points on the physical link to limit the speed of the sliced packets to be transmitted on the physical link.
  • the rate of the network element at the sending end has been limited according to the bandwidth of each rate-limiting point. Therefore, when the sliced packets after the rate limit are actually transmitted on the physical link, the bandwidth requirements of each rate limit point have been met.
  • the network element at the sending end directly obtains the bandwidth of the speed-limiting point itself, and performs speed-limiting based on the obtained bandwidth. There is no need to manually determine the speed-limiting bandwidth of each speed-limiting point, which reduces labor costs, simplifies the process of obtaining the bandwidth for speed-limiting, and improves the obtaining efficiency of the bandwidth for speed-limiting.
  • the speed limit point information further includes: the number of overhead bits of the speed limit point.
  • the process of performing rate-limiting on the slice packets to be transmitted on the physical link includes: for each speed-limiting point of the multiple speed-limiting points , and the difference between the bandwidth of the speed limit point and the bandwidth corresponding to the number of overhead bits is used as the threshold bandwidth; the passing speed per unit time of the sliced packets to pass through the speed limit point is controlled to be less than the threshold bandwidth of the speed limit point.
  • the bandwidth corresponding to the number of overhead bits is the product of the number of overhead bits and the preset packet transmission rate.
  • the network element at the sending end may also update the speed limit point information of at least one speed limit point among the multiple speed limit points on the physical link, and based on the updated bandwidths of the multiple speed limit points, Limit the rate of sliced packets to be transmitted on the physical link.
  • the first method when the information of any speed limit point in the multiple speed limit points changes, obtain the speed limit point information of the any speed limit point; the sending end network element is in any one of the multiple speed limit points When the speed limit point information changes, obtain the speed limit point information of any speed limit point. In this way, it can be ensured that the network element at the sending end can obtain the information of the changed speed limit point in real time, and ensure the timely and accurate speed limit.
  • the speed limit point information of the multiple speed limit points is periodically acquired.
  • the network element at the sending end can obtain the speed limit point information of multiple speed limit points on the physical link in various ways.
  • the process of acquiring the speed limit point information of multiple speed limit points on the physical link by the network element at the sending end includes: the network element at the sending end receives the monitoring information sent by the monitoring device, and the monitoring device communicates with the monitoring device. At least one physical link in the plurality of physical links corresponds.
  • the monitoring information includes speed limit point information of the speed limit point monitored by the monitoring device on the corresponding physical link.
  • each of the multiple physical links includes at least one hardware device, and the hardware device where the speed limit point is located in the at least one hardware device is the monitoring device.
  • the speed limit point monitored by any monitoring device on the corresponding physical link includes the speed limit point located on the any monitoring device and/or the speed limit point connected to the any monitoring device; the receiving monitoring device sends the speed limit point.
  • the process of monitoring information includes: receiving the monitoring information sent by the monitoring device on each physical link.
  • the process of acquiring the speed limit point information of multiple speed limit points on the physical link by the network element at the sending end includes: receiving a bandwidth setting instruction.
  • the bandwidth setting instruction includes speed limit point information of one or more speed limit points in the plurality of physical links.
  • the network element at the sending end has or is connected to an input device, and the user inputs a bandwidth setting instruction through the input device.
  • the network element at the sending end receives the bandwidth setting instruction.
  • the bandwidth setting instruction includes speed limit point information of one or more speed limit points that can be acquired by the user.
  • the speed limit point information is usually speed limit point information of a speed limit point with a fixed bandwidth. In this way, the user usually only needs to set the speed limit point information once, reducing user operations.
  • the speed limit point information obtained by the network element at the sending end includes the speed limit point information reported by the monitoring device and the speed limit point information set through the bandwidth setting instruction.
  • the speed limit point information obtained by the network element at the receiving end may be duplicated.
  • the network element at the sending end may perform deduplication processing on the received speed limit point information to obtain the deduplicated speed limit point information. For example, for the same speed limit point, if multiple speed limit point information is received, obtain the latest speed limit point information in the multiple speed limit point information (that is, the speed limit point information whose setting time or sending time is closest to the current moment) ) as the speed limit point information after deduplication, so that the timeliness and validity of the speed limit point information can be guaranteed.
  • Redundant reporting of the speed limit point information can reduce the risk of missing speed limit point information, and improve the reliability of the speed limit point information received by the network element at the sending end.
  • the communication system also supports a bandwidth prediction function.
  • Flexible rate limiting can be achieved by predicting changes in bandwidth in advance.
  • the bandwidth prediction function includes the following steps: receiving bandwidth prediction information sent by a prediction device.
  • the prediction device corresponds to at least one physical link of the plurality of physical links.
  • the bandwidth prediction information includes predicted bandwidth.
  • the predicted bandwidth is the changed bandwidth of a certain speed limit point predicted by the prediction device after a certain speed limit point monitored on the corresponding physical link satisfies the bandwidth change condition. Based on the predicted bandwidth, the rate of the sliced packets to be transmitted through the certain rate-limiting point among the sliced packets to be transmitted on the physical link is limited.
  • the prediction device may repeatedly send the bandwidth prediction information to ensure that the network element at the sending end can receive the bandwidth prediction information and improve the reliability of reporting the bandwidth prediction information.
  • the prediction device predicts the bandwidth that will take effect in advance, and notifies the network element at the sending end.
  • the network element at the sending end responds and adjusts the bandwidth in the network element in advance, thereby realizing lossless bandwidth switching.
  • it can effectively prevent the loss of sliced packets caused by the failure of the rate limit, thereby avoiding the phenomenon of random packet loss.
  • the process of speed-limiting the sliced packets to be transmitted on the physical link based on the speed-limiting point information of multiple speed-limiting points on the physical link includes: acquiring the attachment relationship information, the The attachment relationship information is used to describe the connection relationship of each speed limit point on the physical link.
  • the attachment relationship information is established based on the identification of each speed limit point on the physical link and the speed limit point type of each speed limit point. Based on the bandwidths of multiple rate-limiting points on the physical link, and according to the connection relationship of each rate-limiting point in the attachment relationship information, multi-level rate-limiting is performed on the sliced packets to be transmitted on the physical link.
  • multi-level rate-limiting is performed on the sliced packets to be transmitted on the physical link.
  • the process includes: for each speed limit point recorded in the attachment relationship information, the difference between the bandwidth of the speed limit point and the bandwidth corresponding to the number of overhead bits is taken as the threshold bandwidth. It is controlled that the passing speed of the sliced packets to pass through the speed limit point per unit time is less than the threshold bandwidth of the speed limit point.
  • the physical link includes m member links, where m is a positive integer.
  • the speed limit point information further includes: the identification of the speed limit point and the speed limit point type.
  • the speed limit point type includes main road speed limit point or branch road speed limit point.
  • the speed limit point of the main road is a speed limit point jointly passed by the m member links
  • the speed limit point of the branch road is a speed limit point passed by the m member links respectively.
  • the attachment relationship information of the physical link includes: m first branches corresponding to the m member links (for example, m first branches correspond to m member links one-to-one), and m first branches corresponding to the m member links A second branch connected by the first branch; then each first branch is used to record the identification of the speed limit point where the speed limit point type on the corresponding member link is the branch speed limit point, and the second branch is used to record
  • the speed limit point type on the physical link is an identifier of the speed limit point of the main road speed limit point.
  • the identifiers of the speed limit points recorded by the branch may be randomly arranged.
  • the identifications of the speed limit points on the branch may be arranged according to a preset rule. For example, the identification of the speed limit point on the branch satisfies at least one of the following conditions:
  • the identifiers of the two rate limiting points are arranged in the branch in descending order of the corresponding number of overhead bits. In this way, when speed limiting is performed, the bandwidth and the number of overhead bits of the previous speed limiting point are preferentially used for speed limiting, so that the sliced packets after the speed limiting reach the latter speed limiting point. To meet the speed requirements, reduce the passing delay of sliced packets at the next speed limit point.
  • the identifiers of the two speed limit points are arranged in the branch according to the order of the corresponding bandwidth from small to large. In this way, when speed limiting is performed, the bandwidth and the number of overhead bits of the previous speed limiting point are preferentially used for speed limiting, so that the sliced packets after the speed limiting reach the latter speed limiting point. To meet the speed requirements, reduce the passing delay of sliced packets at the next speed limit point.
  • each first branch usually records the identification of all speed limit points on the corresponding member link whose type is the speed limit point of the branch; the second branch usually records all the speed limit points on the physical link.
  • the speed limit point type is the identification of the speed limit point of the main road speed limit point.
  • the identification of the speed limit points recorded by the branch can also be integrated, and the identification of one speed limit point among the speed limit points to be integrated can be retained.
  • the number of identifications of the speed limit points recorded in each first branch is less than or equal to the total number of identifications of the speed limit points whose type is the speed limit point of the branch road on the corresponding member link; the speed limit recorded by the second branch
  • the number of identifiers of the points is less than or equal to the total number of identifiers of the speed limit points of the speed limit point type on the physical link of the main road speed limit point.
  • the network element at the sending end can delete the two speed limits recorded in the branch
  • the identification of the speed limit point with the larger bandwidth among the points gets the updated branch.
  • the network element at the sending end can delete the two limit points recorded in the branch. The identification of the speed limit point with the smaller number of overhead bits among the speed points obtains the updated branch.
  • the network element at the sending end can delete the two limit points recorded in the branch.
  • the identification of a speed limit point in the speed point gets the updated branch.
  • the speed limit based on the aforementioned updated branch can achieve the same effect as the speed limit based on the unupdated branch, but the updated branch records have fewer speed limit point identifiers and lower complexity.
  • the network element at the sending end of the embodiment of the present application also supports the flow control back pressure method.
  • Flow control back pressure refers to the flow control of sliced packets through feedback.
  • the flow control backpressure process includes: for each speed limit point recorded in the attachment relationship information, when the length of the sliced packet to pass through the speed limit point and the overhead bit of the speed limit point The bandwidth corresponding to the sum of the numbers is greater than the bandwidth of the speed limit point. The bandwidth corresponding to the sum of the length of the sliced packets to pass through the speed limit point and the number of overhead bits of the speed limit point is reduced by discarding the service packets, and the service packets are used for slicing to obtain the sliced packets.
  • a communication device in a second aspect, includes: a processing chip, the processing chip includes: a processing circuit and a communication interface, the processing circuit is used for executing any one of the communication methods in the first aspect; the communication interface is used for the processing circuit to communicate with other devices.
  • the processing circuit may be a processing chip or a field programmable gate array (Field Programmable Gate Array, FPGA).
  • a communication device in a third aspect, includes at least one module, and the at least one module can be used to implement the communication method provided by the first aspect or various possible implementation manners of the first aspect.
  • the present application provides a communication system.
  • the communication system includes a sending end network element, a receiving end network element, and a plurality of physical links bundled by aggregating PLA through physical links, and the sending end network element includes the communication device described in the second aspect or the third aspect.
  • the present application provides a computer device including a processor and a memory.
  • the memory stores computer instructions; the processor executes the computer instructions stored in the memory, so that the computer device performs the first aspect or the methods provided by various possible implementations of the first aspect.
  • the present application provides a computer-readable storage medium, where computer instructions are stored in the computer-readable storage medium, and the computer instructions instruct a computer device to execute the method provided in the first aspect or various possible implementations thereof.
  • the present application provides a computer program product comprising computer instructions stored in a computer-readable storage medium.
  • the processor of the computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the methods provided by various possible implementations of the first aspect above.
  • a chip in an eighth aspect, may include programmable logic circuits for implementing a communication method as in any of the first aspects when the chip is in operation.
  • the network element at the sending end uses the obtained bandwidths of multiple speed limiting points on the physical link to limit the speed of the sliced packets to be transmitted on the physical link.
  • the rate of the network element at the sending end has been limited according to the bandwidth of each rate-limiting point. Therefore, when the sliced packets after the rate limit are actually transmitted on the physical link, the bandwidth requirements of each rate limit point have been met.
  • the network element at the sending end directly obtains the bandwidth of the speed-limiting point itself, and performs speed-limiting based on the obtained bandwidth. There is no need to manually determine the speed-limiting bandwidth of each speed-limiting point, which reduces labor costs, simplifies the process of obtaining the bandwidth for speed-limiting, and improves the obtaining efficiency of the bandwidth for speed-limiting.
  • FIG. 1 is a schematic structural diagram of a communication system involved in a communication method provided by the present application.
  • FIG. 2 is a schematic diagram of a communication principle of a communication system involved in a communication method provided by the present application
  • FIG. 3 is a schematic flowchart of a communication method provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a schematic attachment relationship provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of the attachment relationship information of the physical link shown in FIG. 1;
  • Fig. 6 is the connection network diagram of the speed limiter corresponding to the attachment relationship information of the physical link shown in Fig. 5;
  • FIG. 7 is a schematic structural diagram of a communication device provided by an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of another communication apparatus provided by an embodiment of the present application.
  • FIG. 1 is a schematic structural diagram of a communication system involved in a communication method provided by an embodiment of the present application.
  • the communication system includes: a sending end network element 101 , a receiving end network element 102 and an aggregated link.
  • the aggregated link includes multiple physical links that are bundled.
  • FIG. 1 assumes that the aggregated link includes two physical links, namely physical link 1031 and physical link 1032 .
  • an aggregated link corresponds to a logical interface, which is also called a link aggregation interface.
  • a link corresponding to each physical interface (also called a physical port or a hardware interface) that constitutes a link aggregation interface is called a physical link.
  • the physical link may refer to a direct path connecting two network elements (ie, there is no intermediate network element between the two network elements) or an indirect path (ie, there is an intermediate network element between the two network elements).
  • the physical link is usually a physical link.
  • a physical link between the sending end network element 101 and the receiving end network element 102 is a direct or indirect path between a physical interface on the sending end network element 101 and a physical interface on the receiving end network element 102 .
  • the network element 101 at the sending end is configured to slice the service packet to obtain a plurality of sliced packets.
  • the aggregated link is used to transmit the multiple sliced packets.
  • the network element 102 at the receiving end is configured to receive and sort a plurality of sliced packets, and then reassemble the sorted sliced packets to obtain a service packet.
  • the network elements may be routers, switches, or other types of network elements.
  • the aggregated link includes one or more types of physical links.
  • the aggregated link includes: an Ethernet link, a microwave link, an Optical Transport Network (OTN) link, and an improved OTN link (also called an enhanced OTN link or an evolved OTN link).
  • OTN Optical Transport Network
  • the physical links in the aggregated link may be deployed by the same operator, or may be deployed by different operators.
  • the aggregated link includes the microwave link as an example.
  • PLA technology using microwave links can be divided into ordinary PLA (also known as traditional PLA) technology, enhanced PLA (Enhanced Physical Link Aggregation, EPLA) technology and Super Dual Band (Super Dual Band, SDB) technology and other types.
  • the aggregated link includes two microwave links.
  • the aggregated link includes at least three microwave links; in the SDB technology, the aggregated link includes two microwave links in different frequency bands.
  • microwave links in conventional frequency bands and microwave links in E-BAND (E-band) are included.
  • the regular frequency band is 7-38 gigahertz (GHz) and the E-BAND is 71-86GHz.
  • FIG. 2 is a schematic diagram of a communication principle of a communication system involved in a communication method provided by an embodiment of the present application.
  • each physical link is configured with one or more member links (it can also be regarded as dividing the physical link into one or more member links), then an aggregated link includes multiple members
  • the number of member links of an aggregated link is greater than or equal to the number of physical links.
  • the member links belonging to the same physical link together form a physical interface of the link aggregation interface, different member links belonging to the same physical link use different communication interfaces respectively, and at least one member link belonging to the same physical link One-to-one correspondence with at least one communication interface.
  • the communication interface can be an air interface or a wired interface, and the wired interface can be an optical interface or an electrical interface.
  • the member link uses an air interface for communication, the member link includes a wireless link connected to the air interface; when a member link uses a wired interface for communication, the member link includes a communication cable connected to the wired interface.
  • the network element at the sending end is used to perform a packet slicing process on service packets to obtain multiple sliced packets, and then perform a packet distribution process on the multiple sliced packets, so as to Each slice packet is distributed to multiple member links in the aggregated link; the network element at the receiving end is used to perform the slice collection process to collect multiple slice packets transmitted on the member links; A slice caching process is performed for each sliced packet to sort the plurality of sliced packets; and then a packet reorganization process is performed to reorganize the sorted sliced packets to obtain service packets.
  • the aggregated link includes n member links, which are member link 1 to member link n, respectively, and n ⁇ 2.
  • the sending end network element 101 performs the packet slicing process and the packet distribution process on the service packets, it distributes the sliced packets 1, 3, 7 and 8 to the member link 1, and the sliced packets 2, 4, 5 and 8. 6 is distributed to the member link n; the network element 102 at the receiving end obtains the service packet composed of the slice packets 1-8 by executing the slice collection process, the slice cache process and the packet reorganization process.
  • a rate limit point is relative to the sliced packets sent by the network element at the sending end, and is a point that has a speed-limiting effect on the sliced packets sent by the network element at the sending end.
  • a rate limit point can be a physical interface on a backplane, a single board, or a device.
  • the physical interface can be a connector, an optical fiber connector, a single-board plug, or an Ethernet interface.
  • the single board may include an IF board.
  • the speed limit point can also be a communication interface, such as an air interface or a wired interface.
  • the rate-limiting point usually encapsulates additional overhead bits (or overhead bytes) in the sliced packet.
  • speed limit points with variable bandwidth speed limit points with fixed bandwidth.
  • the bandwidth of the bandwidth-variable rate-limiting point is easily affected by external environments, such as weather.
  • the speed limit point is an air interface (eg, a microwave air interface).
  • the bandwidth of the speed-limiting point with a fixed bandwidth is not affected by the external environment, and in the communication system, the bandwidth of the speed-limiting point is not changed after the aggregation link is deployed.
  • an Ethernet physical port has a bandwidth of 1 Gpbs (gigabits per second).
  • the following description is given by taking FIG. 1 as an example.
  • the sliced packets generated by the network element 101 at the sending end pass through the backplane 1, the board 1, the board 2 of the network element 101 at the sending end, and the backplane 2 of the network element 102 at the receiving end in sequence, so as to be transmitted to the receiving end network element 102.
  • End network element 102 End network element 102 .
  • the backplane 1 is integrated on the network element 101 at the sending end, and can be integrated with the network element 101 at the sending end.
  • the backplane 2 is integrated on the network element 102 at the receiving end, and may have an integrated structure with the network element 102 at the receiving end.
  • the backplane 1 has a connector A
  • the single board 1 has a connector B for connecting with the backplane
  • air interfaces C1 and C2 for communicating with the network element 102 at the receiving end.
  • the connector A, the connector B, the air interface C1 and the air interface C2 respectively exert a rate limiting effect on the sliced packets sent by the network element 101 at the sending end.
  • the speed limit points of the physical link 1031 include: connector A, connector B, air interface C1 and air interface C2.
  • the physical link 1031 includes two member links, respectively a member link passing through connector A, connector B and air interface C1, and a member link passing through connector A, connector B and air interface C2.
  • the slice packet generated by the network element at the sending end is transmitted to the network element 102 at the receiving end through the network element 101 at the sending end, the slave device 1, and the slave device 2 in sequence.
  • the sending end network element 101 has an Ethernet interface D
  • the slave device 1 has an Ethernet interface E for connecting with the sending end network element 101 and an air interface F for communicating with the receiving end network element 102 .
  • the Ethernet interface D, the Ethernet interface E, and the air interface F respectively limit the rate of the sliced packets sent by the network element 101 at the sending end.
  • the rate limiting points of the physical link 1032 include: Ethernet interface D, Ethernet interface E, and air interface F
  • the physical link 1032 includes a member link, that is, a member link passing through Ethernet interface D, Ethernet interface E, and air interface F.
  • two rate-limiting points with the same bandwidth and the same number of overhead bits have the same rate-limiting effect (that is, the rate-limiting effect of the sliced packets passing through the two rate-limiting points is the same as the rate-limiting effect after passing through the two rate-limiting points).
  • the speed limit effect after any one of the speed limit points is the same), and the two can be regarded as the same speed limit point.
  • the two connectors after connector A and connector B are connected (eg, plugged), the two connectors have the same rate limiting effect on the sliced packets sent by the network element 101 at the sending end (that is, the sliced packets pass through the two connectors).
  • the speed-limiting effect after passing through either of the two connectors is the same).
  • both can be regarded as the same connector, and can be regarded as the same speed limit point on the physical link.
  • the Ethernet interface D and the Ethernet interface E are connected through an Ethernet port connection. Since the Ethernet port connection usually does not have a rate limiting function, the two Ethernet interfaces limit the rate of the sliced packets sent by the NE 101 at the sending end. The functions are the same (that is, the rate limiting effect of sliced packets passing through Ethernet interface D, Ethernet interface connection, and Ethernet interface E is the same as the rate limiting effect after passing through either Ethernet interface D or Ethernet interface E). In this way, the two Ethernet interfaces can be regarded as the same Ethernet interface, and can be regarded as the same speed limit point on the physical link. Treating the speed limit point with the same speed limit function of the physical link (for example, the adjacent speed limit point with the same speed limit function) as a speed limit point can reduce the complexity of the acquired physical link and improve the performance of the physical link. Link management efficiency.
  • the backplane 2 in FIG. 1 has a connector, the single board 2 has a connector and an air interface, the network element 102 at the receiving end has an Ethernet interface, and the slave device 2 has an Ethernet interface and an air interface.
  • Figure 1 assumes that these physical interfaces and communication interfaces do not limit the rate of the sliced packets sent by the network element at the sending end, that is, they are not rate-limiting points for the sliced packets. These physical and communication interfaces are therefore not shown in FIG. 1 . However, in actual implementation, on the physical link, there may be a speed limit point on either the side where the network element at the sending end or the side where the network element at the receiving end is located, which is not limited in this embodiment of the present application.
  • the bandwidth of sliced packets allowed to pass through each rate limit point is affected by the following factors:
  • each rate limit point will add some overhead bits to the sliced packet. Different rate limit points have different numbers of overhead bits added. . The same speed limit point, the same number of overhead bits added.
  • the length of the sliced packet For different sliced scenarios, the length of the sliced packet may be different, for example, the length of the sliced packet may be 64 bytes or 1518 bytes.
  • the payloads carried in the packets of different lengths passing through the same rate limiting point within a unit duration are different.
  • the payload refers to the data actually carried by the packet.
  • the payload of the sliced packet passing through a rate limit point in a unit time is called the maximum payload bandwidth, and the maximum payload bandwidth is positively related to the length of the sliced packet. That is, the longer the length of the sliced packet, the greater the maximum payload bandwidth, and the higher the benefit of sliced packet transmission.
  • the bandwidth used for speed limit of the member link is usually determined by considering the worst transmission situation.
  • the bandwidth of the member link is determined in the following manner: for each member link, the staff is based on the bandwidth of each speed limit point on the member link, the overhead bit, and the minimum length of the sliced packet passing through the speed limit point. and the bandwidth of adjacent speed limit points to determine the speed limit bandwidth of each speed limit point; and then determine the minimum value of the speed limit bandwidth of each speed limit point as the bandwidth of the member link.
  • the process of obtaining the rate-limiting bandwidth of each rate-limiting point is complicated, and the labor cost is high. As a result, the acquisition process of the bandwidth used for speed limitation is complicated, and the acquisition efficiency is low.
  • FIG. 3 is a schematic flowchart of a communication method provided by an embodiment of the present application.
  • the communication method is applied to the communication system shown in FIG. 1 or FIG. 2 .
  • the method can reduce labor costs, simplify the process of obtaining bandwidth for speed reduction, and improve the efficiency of obtaining bandwidth used for speed reduction.
  • the processing actions performed by the network element at the sending end are the same.
  • a physical link is used as an example for description below, and the processing actions for other physical links refer to the processing actions for the one physical link.
  • the method includes the following steps.
  • the network element at the sending end acquires speed limit point information of multiple speed limit points on the physical link, where the speed limit point information includes bandwidth (ie, actual bandwidth).
  • the network element at the sending end may acquire the speed limit point information of multiple speed limit points on the physical link in various ways.
  • the network element at the sending end receives the speed limit point information reported by the monitoring device.
  • the monitoring device corresponds to at least one physical link in the aggregated link, and the monitoring device is used to monitor the speed limit point information of the corresponding physical link speed limit point.
  • the process of the sending end network element receiving the speed limit point information reported by the monitoring device includes: the monitoring device sends monitoring information to the sending end network element, where the monitoring information includes the speed limit point information monitored by the monitoring device on the corresponding physical link. Speed limit point information; correspondingly, the sending end device receives the monitoring information sent by the monitoring device.
  • the monitoring information further includes an information identifier.
  • the information identifier is used to uniquely identify the monitoring information, which is used to distinguish it from other control information transmitted in the network element at the sending end.
  • the monitoring information may further include a priority, and the monitoring device may determine the order of sending the monitoring information according to the priority of the monitoring information. For example, the priority of the monitoring information is the highest priority of the communication system.
  • each physical link in the aggregated link includes at least one hardware device, the at least one hardware device includes the aforementioned monitoring device, and the physical link corresponding to the monitoring device is the physical link where the monitoring device is located.
  • the monitoring device on each physical link sends monitoring information to the network element at the sending end; correspondingly, the network element at the sending end receives the monitoring information sent by the monitoring device on each physical link.
  • the hardware device is a device capable of sending information to the network element at the sending end, which may be a backplane, a single board, a slave device, or other structures.
  • the hardware device where the speed limit point in the at least one hardware device is located is a monitoring device, and the speed limit point monitored by any monitoring device on the corresponding physical link includes the speed limit point located on the any monitoring device.
  • the backplane 1 and the single board 1 may both be monitoring devices on the physical link 1031 , and the slave device 1 is the monitoring device on the physical link 1032 .
  • the monitoring information sent by the backplane 1 includes: the speed limit point of the speed limit point A
  • the monitoring information sent by the board 1 includes: the speed limit information of the speed limit points B, C1 and C2; the monitoring information sent by the slave device 1 includes: the speed limit point information of the speed limit points E and F.
  • the monitoring information sent by the backplane 1 may include: speed limit points B, C1 and C2.
  • the monitoring device sent by the backplane 1 The information includes: speed limit point information of speed limit points A, B, C1 and C2; the monitoring information sent by board 1 includes: speed limit point information of speed limit points A, B, C1 and C2; the monitoring information sent from device 1 The information includes: speed limit information of speed limit points D, E and F.
  • the designated hardware device in the at least one hardware device is a monitoring device, and the speed limit point monitored by the monitoring device on the corresponding physical link includes one or more speed limit points on the corresponding physical link Speed limit points, such as all speed limit points on that physical link.
  • the monitoring information sent by the board 1 includes: the speed limit information of the speed limit points A, B, C1 and C2; the monitoring information sent by the slave device 1 includes: the speed limit point information of the speed limit points D, E and F.
  • a physical link includes a sender link, which is determined along the direction away from the sender network element from the first speed limit point (such as a physical interface) to the last speed limit point (such as an air interface). ), since the link at the sending end is close to the network element at the sending end, and the hardware devices outside the link at the sending end are far away from the network element at the sending end, therefore, the specified hardware device (that is, the monitoring device) on the sender link.
  • the designated hardware device is the hardware device on the link of the sending end, the rapid reporting of monitoring information can be realized, and the transmission delay of the monitoring information can be reduced.
  • the monitoring device is a structure with a communication function, which can send information to the network element at the sending end. Since the traditional backplane does not have a communication function, if the backplane needs to be configured as a monitoring device, the backplane is an intelligent backplane with a communication function.
  • the monitoring device is located outside the aggregated link, and is used to monitor the speed limit point on one or more physical links in the aggregated link, and the monitoring device performs data from the monitored physical link from the speed limit point.
  • the speed limit point information of each speed limit point is collected on the hardware device, and the collected speed limit point information is sent to the sender device.
  • the network element at the sending end obtains the speed limit point information through manual configuration.
  • the process of receiving the speed limit point information reported by the monitoring device by the network element at the sending end includes: receiving a bandwidth setting instruction, where the bandwidth setting instruction includes the speed limit point information of one or more speed limit points in the multiple physical links of the aggregated link. .
  • the network element at the sending end has or is connected to an input device, the user inputs a bandwidth setting instruction through the input device, and accordingly, the network element at the sending end receives the bandwidth setting instruction.
  • the bandwidth setting instruction includes speed limit point information of one or more speed limit points that can be acquired by the user.
  • the speed limit point information is usually speed limit point information of a speed limit point with a fixed bandwidth. In this way, the user usually only needs to set the speed limit point information once, reducing user operations.
  • the first speed limit point on the physical link is usually the physical interface of the sending-end NE, as shown in Figure 1, the connector A of the backplane 1 on the sending-end NE, and the sending-end NE Ethernet interface D. Therefore, in actual implementation, in addition to receiving the speed limit point information of the first speed limit point reported by the monitoring device, or receiving the manually configured speed limit point information of the first speed limit point, the sender network element can also Directly obtain the speed limit information of its own physical interface.
  • the speed limit point information obtained by the network element at the sending end includes the speed limit point information reported by the monitoring device and the speed limit point information set through the bandwidth setting instruction.
  • the speed limit point information obtained by the network element at the sending end may be duplicated.
  • the speed limit point monitored by any monitoring device on the corresponding physical link includes the speed limit point located on any monitoring device and the speed limit point that is related to any monitoring device
  • the speed limit points monitored by the monitoring devices located on the same physical link may overlap, so that the reported speed limit point information of the same speed limit point is repeated.
  • the speed limit point information reported by the monitoring device and the manually set speed limit point information are repeated.
  • the network element at the sending end may perform deduplication processing on the received speed limit point information to obtain the deduplicated speed limit point information. For example, for the same speed limit point, if multiple speed limit point information is received, obtain the latest speed limit point information in the multiple speed limit point information (that is, the speed limit point information whose setting time or sending time is closest to the current moment) ) as the speed limit point information after deduplication, so that the timeliness and validity of the speed limit point information can be guaranteed.
  • Redundant reporting of the speed limit point information can reduce the risk of missing speed limit point information, and improve the reliability of the speed limit point information received by the network element at the sending end.
  • the network element at the sending end limits the rate of the sliced packets transmitted on the physical link based on the bandwidths of multiple rate limiting points on the physical link.
  • the rate limit process also needs to consider the number of overhead bits added by the rate limit point. Then, for each speed limit point of the multiple speed limit points, the network element at the sending end can use the difference between the bandwidth of the speed limit point and the bandwidth corresponding to the number of overhead bits as the threshold bandwidth, and control the slice report to pass through the speed limit point.
  • the passing speed of the text per unit time is less than the threshold bandwidth of the speed limit point.
  • the bandwidth corresponding to the number of overhead bits is the product of the number of overhead bits and the preset packet transmission rate.
  • the network element at the sending end may first determine the connection relationship of each speed-limiting point on the physical link, and then perform speed-limiting based on this. Then, based on the rate-limiting point information of multiple rate-limiting points on the physical link, the process of speed-limiting the sliced packets transmitted on the physical link may include:
  • the network element at the sending end acquires the attachment relationship information, where the attachment relationship information is used to describe the connection relationship of each speed limit point on the physical link.
  • the sending-end network element After acquiring the speed-limiting point information of the multiple speed-limiting points, the sending-end network element establishes a connection relationship diagram based on the acquired speed-limiting point information.
  • the speed limit point information may further include the speed limit point identification and speed limit point type, and the speed limit point identification may include the identification of the device where the speed limit point is located and/or the speed limit point number.
  • the identification of the device may include: an Internet Protocol (Internet Protocol, IP) address and/or a network element identification (NE Identity, NEID).
  • IP Internet Protocol
  • NE Identity network element identification
  • the attachment relationship information is established based on the identification of each speed limit point on the physical link and the speed limit point type of each speed limit point.
  • the aforementioned attachment relationship information may be represented by a relationship diagram or a relationship table.
  • the aforementioned speed limit point types include main road speed limit points or branch road speed limit points.
  • a physical link includes m member links, m is a positive integer, the speed limit point of the main road is the speed limit point jointly passed by the m member links, and the speed limit point of the branch road is the speed limit point passed by the m member links respectively speed limit point.
  • the physical link includes two member links.
  • the two member links are the first member link passing through the speed limit points A, B, and C1, and the two member links are respectively The second member link passing through the speed limit points A, B and C2, the main road speed limit points are the speed limit points A and B, the branch road speed limit points are the speed limit point C1 passed by the first member link respectively, and The speed limit point C2 that the second member link passes through.
  • FIG. 4 is a schematic diagram of a schematic attachment relationship provided by an embodiment of the present application.
  • the attachment relationship information of the physical link includes: m first branches X corresponding to m member links (for example, m first branches X correspond one-to-one with m member links), and m first branches X corresponding to m member links A second branch Y to which branch X is connected.
  • Each first branch X is used to record the speed limit point identification x1 whose speed limit point type on the corresponding member link is the branch speed limit point, and the number of speed limit point identifications recorded on different first branches X Can be the same or different.
  • the second branch Y is used to record the speed limit point identifier y1 of the speed limit point type of the main road speed limit point on the physical link.
  • the second branch Y is located after the first branch X, so when the speed limit is performed based on the attachment relationship diagram, the speed limit is first performed based on the first branch, then the speed limit is performed based on the second branch, and finally the speed limit is completed.
  • Sliced packets are output from the sender NE.
  • FIG. 5 is a schematic diagram of the attachment relationship information of the physical link 1031 shown in FIG. 1 .
  • the physical link 1031 includes a first member link passing through the speed limit points A, B and C1, and a second member link passing through the speed limit points A, B and C2 respectively.
  • the corresponding attachment relationship information includes: two first branches corresponding to the two member links, and a second branch Y connected to the two first branches.
  • the first branches corresponding to the first member link and the second member link are the first branches X1 and X2, respectively, the first branch X1 is used to record the identification C1 of the speed limit point on the first member link, and the first branch X1 A branch X2 is used to record the identification C2 of the speed limit point on the second member link, and the second branch Y is used to record the identification A and B.
  • the identifiers of the speed limit points recorded by the branch may be randomly arranged.
  • the identifications of the speed limit points on the branch may be arranged according to a preset rule. For example, when the NE at the sending end limits the rate of sliced packets in the NE, the actual rate limiting effect is not only affected by the bandwidth of the rate-limited point, but also by the number of overhead bits of the rate-limited point. The number of overhead bits of the speed point and the bandwidth are used to arrange the identification of the speed limit point in the branch.
  • the identification of the speed limit point on this branch satisfies at least one of the following conditions:
  • the identifiers of the two rate limiting points are arranged in the branch in descending order of the corresponding number of overhead bits. Since the length of the sliced packets passing through the speed limit point remains unchanged and the bandwidth of the speed limit point remains unchanged, the actual speed limit effect of the speed limit point is determined by the number of overhead bits added. The greater the restriction on the transmission rate of sliced packets. Therefore, when the bandwidths of the two speed limit points are the same and the number of overhead bits is different, the identifier of the corresponding speed limit point with the larger number of overhead bits is ranked first.
  • the bandwidth and the number of overhead bits of the previous speed limiting point are preferentially used for speed limiting, so that the sliced packets after the speed limiting reach the latter speed limiting point.
  • the identifiers of the two speed limit points are arranged in the branch according to the order of the corresponding bandwidth from small to large. Since the length of the sliced packet passing through the rate-limiting point remains unchanged, and the number of overhead bits of the rate-limiting point remains unchanged, the actual rate-limiting effect of the rate-limiting point is determined by its bandwidth. The higher the rate limit. Therefore, when the number of overhead bits of the two speed limit points is the same and the bandwidth is different, the identifier of the corresponding speed limit point with the smaller bandwidth is ranked first.
  • the bandwidth and the number of overhead bits of the previous speed limiting point are preferentially used for speed limiting, so that the sliced packets after the speed limiting reach the latter speed limiting point.
  • each first branch usually records the identification of all speed limit points on the corresponding member link whose type is the speed limit point of the branch; the second branch usually records all the speed limit points on the physical link.
  • the speed limit point type is the identification of the speed limit point of the main road speed limit point.
  • the identification of the speed limit points recorded by the branch can also be integrated, and the identification of one speed limit point among the speed limit points to be integrated can be retained.
  • the number of identifications of the speed limit points recorded in each first branch is less than or equal to the total number of identifications of the speed limit points whose type is the speed limit point of the branch road on the corresponding member link; the speed limit recorded by the second branch
  • the number of identifiers of the points is less than or equal to the total number of identifiers of the speed limit points of the speed limit point type on the physical link of the main road speed limit point.
  • the identification of the speed limit points in the branch can be integrated based on the number of overhead bits of the speed limit point and the bandwidth.
  • the integrated speed limit point identification mainly includes the following three categories:
  • the first category is the identification of two rate-limiting points with the same number of overhead bits and the same bandwidth.
  • the signs of the two speed limit points are integrated into one speed limit point sign, and finally the updated branch.
  • Speed limiting based on this branch can achieve the same effect as speed limiting based on the branch before the update.
  • the updated branch records have fewer identification numbers and lower complexity.
  • the integration process includes: for each branch in the first branch and the second branch, when the number of overhead bits of the two speed limit points is the same and the bandwidth is the same, deleting the two speed limit points recorded in the branch The identification of a speed limit point in the branch gets updated.
  • the second type is the identification of two rate-limiting points with the same number of overhead bits and different bandwidths.
  • the speed limit point with small bandwidth can already achieve effective speed limit compared with the speed limit point with large bandwidth, the speed limit point with small bandwidth can be reserved. id, and finally get the updated branch.
  • Speed limiting based on this branch can achieve the same effect as speed limiting based on the branch before the update.
  • the updated branch records have fewer identification numbers and lower complexity.
  • the integration process includes: for each branch in the first branch and the second branch, when the number of overhead bits of the two speed limit points is the same and the bandwidth is different, deleting the two speed limit points recorded in the branch The identification of the speed limit point with the larger bandwidth in the middle gets the updated branch.
  • the third category is the identification of two rate-limiting points with different overhead bits and the same bandwidth.
  • the limit with a small overhead bit can be reserved.
  • the logo of the speed point, and finally the updated branch is obtained.
  • Speed limiting based on this branch can achieve the same effect as speed limiting based on the branch before the update.
  • the updated branch records have fewer identification numbers and lower complexity.
  • the integration process includes: for each branch in the first branch and the second branch, when the number of overhead bits of the two speed limit points is different and the bandwidth is the same, deleting the two speed limit points recorded in the branch The identification of the speed limit point with the smaller number of overhead bits in the middle gets the updated branch.
  • the updated branch structure is concise, which can reduce the complexity of speed limiting based on this branch.
  • the foregoing speed limit point information also includes the number of overhead bits.
  • the network element at the sending end can arrange and/or integrate the identifications of the speed limit points on the branch by using the number of overhead bits and the bandwidth in the obtained speed limit point information.
  • the network element at the sending end Based on the bandwidth of multiple speed-limiting points on the physical link, the network element at the sending end performs multi-level rate-limiting on the sliced packets to be transmitted on the physical link according to the connection relationship of each speed-limiting point in the attachment relationship information. .
  • the identification of each speed limit point recorded in the attachment relationship information corresponds to a bandwidth.
  • the network element at the sending end sequentially limits the rate of the sliced packets to be transmitted on the physical link by using the bandwidth corresponding to the identifier of the speed limit point according to the sequence of the identifiers of the speed limit points of each branch in the attachment relationship information.
  • the speed limit on the branch corresponding to a member link may include one-level speed limit or multi-level speed limit.
  • the multi-level speed limit means that each slice packet passes through at least two levels The bandwidth of the speed limit point will limit it.
  • the rate limit is adopted.
  • the bandwidth of point C1 the bandwidth of rate-limiting point A, and the bandwidth of rate-limiting point B sequentially limit the rate of the sliced packet; if the sliced packet is a sliced packet to pass through the second member link, the rate-limiting point is used.
  • the bandwidth of C2 the bandwidth of rate limiting point A, and the bandwidth of rate limiting point B sequentially limit the rate of the sliced packets.
  • the process of using the bandwidth of the rate limit point to limit the rate of the packet includes: adding the bandwidth of the rate limit point to the bandwidth corresponding to the number of overhead bits. The difference is used as the threshold bandwidth, and the passing speed per unit time of the sliced packets to be passed through the speed limit point is controlled to be less than the threshold bandwidth of the speed limit point.
  • the network element at the sending end establishes a corresponding relationship between the identification of the speed limit point and the bandwidth based on the obtained speed limit point information of multiple speed limit points on the physical link, and stores the corresponding relationship in the in the preset storage space.
  • the identification is used to query the corresponding relationship between the identification of the rate limiting point and the bandwidth, and the packet segmentation is performed based on the bandwidth obtained by the query. speed limit.
  • the attachment relationship information records the identification of the speed-limiting point and the bandwidth corresponding to the identification of each speed-limiting point.
  • the bandwidth corresponding to the identification is used to limit the rate of sliced packets. Since the identifier corresponding to the speed limit point is directly recorded in the connection relationship information, there is no need to query the bandwidth corresponding to the identifier of the speed limit point during the speed limit process, which reduces the complexity of the speed limit process and reduces the speed limit delay.
  • the network element at the sending end may implement multi-level rate limiting for sliced packets by means of hardware rate limiting or software rate limiting.
  • a preset number of hardware speed limiters are pre-configured in the network element at the sending end.
  • the speed limiter is a speed limit circuit or a speed limit chip, and the preset number is usually greater than or equal to the total number of speed limit points of the physical link.
  • the network element at the sending end selects multiple speed limiters from the preset number of speed limiters according to the sequence of the identification of each speed limit point in the connection relationship information, and establishes multiple speed limiters.
  • the multiple speed limiters are in one-to-one correspondence with the identifiers of the multiple speed limit points in the attachment relationship information. The process of establishing a connection network with multiple speed limit points is called hooking.
  • the entrance of the connection network of the multiple speed limiters is connected with the output end of the slice module in the network element at the sending end, and the outlet of the connection network and the network element at the sending end are used to output the slice report to the physical link.
  • the slicing module is used to perform a packet slicing process on service packets.
  • Each rate limiter is configured to use the bandwidth of the corresponding rate limit point to limit the rate of the sliced packets passing through the rate limiter.
  • the speed limit process refers to that the speed limiter controls the passing speed per unit time of the sliced packets passing through the speed limiter to be less than the bandwidth of the corresponding speed limit point, and the passing speed per unit time of the sliced packets can be It is represented by the product of the length of the sliced packet and the preset packet transmission rate. As mentioned above, since the sliced packets that are allowed to pass through each speed limit point are also affected by the number of overhead bits added by the speed limit point, each speed limiter can simulate the actual speed limit situation of the corresponding speed limit point. speed limit.
  • the speed limit process includes: the speed limiter takes the difference between the bandwidth of the speed limit point and the bandwidth corresponding to the number of overhead bits as the threshold bandwidth, and controls the passing speed per unit time of the sliced packets passing through the speed limiter to be less than the corresponding limit.
  • the bandwidth corresponding to the number of overhead bits is the product of the number of overhead bits and the preset packet transmission rate.
  • FIG. 6 is a schematic diagram of the connection network of the speed limiter corresponding to the attachment relationship information of the physical link shown in FIG. 5 .
  • the network element at the sending end establishes a connection network of four speed limiters according to the sequence of the identifiers of the speed limit points in the attachment relationship information.
  • the four speed limiters are respectively speed limiters 1-4.
  • the four speed limiters are in one-to-one correspondence with the signs C1, C2, A and B of the four speed limit points in the attachment relationship information.
  • the speed limiters 1 and 2 corresponding to the speed limit points C1 and C2 are respectively connected with the speed limiter 3 corresponding to the speed limit point A, and the speed limiter 3 is also connected with the speed limiter corresponding to the speed limit point B. 4.
  • the speed limiters 1 and 2 are the entrances connecting to the network, and the speed limiter 4 is the exit connecting to the network.
  • the network element at the sending end allocates multiple software speed limiters (that is, assigns multiple software speed limiters) according to the sequence of the identification of each speed limit point in the attachment relationship information. resources for speed limiting), and establish a connection network of the multiple speed limiters.
  • the multiple speed limiters are in one-to-one correspondence with the identifiers of the multiple speed limit points in the attachment relationship information.
  • the process of establishing a connection network with multiple speed limit points is called hooking.
  • the connection network of the software speed limiter after the connection is completed is the same as the connection network of the hardware speed limiter.
  • connection method and function of each speed limiter please refer to the connection method of the speed limiter in the hardware speed limiter. and function.
  • the aforementioned hardware speed limit mode has higher speed limit reliability, and the speed limiter is easy to manage and maintain.
  • a scheduler needs to be set, and the scheduler can be a hardware scheduler or a software scheduler. It should be a connection point between multiple first branches and one second branch, and the scheduler is used to schedule slice packets from different first branches corresponding to speed limiters, so as to prevent congestion of slice packets.
  • the aforementioned scheduler is a Round Robin (RR) scheduler.
  • the network element at the sending end updates the speed limit point information of at least one speed limit point among the multiple speed limit points on the physical link.
  • Speed limit information for multiple speed limit points may change due to changes in the environment or other factors. For example, if the speed limit point is an air interface, the bandwidth of the speed limit point changes due to changes in the environment where the speed limit point is located, such as changes in weather conditions. For another example, after the hardware device where a certain speed limit point is located fails, the hardware device is replaced. As a result, the certain speed limit point is updated, and the corresponding speed limit point information, such as the bandwidth and the identification of the speed limit point, also changes. The network element at the sending end needs to obtain the changed speed limit point information to ensure the speed limit effect.
  • the network element at the sending end acquires the speed limit point information of the any speed limit point.
  • the process of acquiring the speed-limiting point information of any speed-limiting point by the transmitting-end network element reference may be made to the process of acquiring the speed-limiting point information of a speed-limiting point on the physical link by the transmitting-end network element in the foregoing S301.
  • the speed limit point information of any speed limit point reported by the monitoring device is received, or the speed limit point information of any speed limit point is obtained through manual configuration.
  • the network element at the sending end acquires the speed limit point information of the any speed limit point. In this way, it can be ensured that the network element at the sending end can obtain the information of the changed speed limit point in real time, and ensure the timely and accurate speed limit.
  • the network element at the sending end periodically obtains the speed limit point information of multiple speed limit points.
  • each acquisition cycle for the process of acquiring the speed-limiting point information of multiple speed-limiting points by the transmitting-end network element, refer to the process of acquiring the speed-limiting point information of multiple speed-limiting points on the physical link by the transmitting-end network element in the foregoing S301 .
  • the speed limit point information of the multiple speed limit points reported by the monitoring device is received, or the speed limit point information of the multiple speed limit points is acquired through manual configuration.
  • the network element at the sending end performs rate limiting on the sliced packets to be transmitted on the physical link based on the updated bandwidths of the multiple rate limiting points.
  • the sending-end network element may refer to the rate-limiting process in the foregoing S302 for the process of speed-limiting the sliced packets to be transmitted on the physical link.
  • the network element at the sending end if the network element at the sending end establishes a corresponding relationship between the identification of the speed-limiting point and the bandwidth, the network element at the transmitting end first updates the identification of the speed-limiting point and the bandwidth based on the updated speed-limiting point information of the multiple speed-limiting points. Corresponding relationship of bandwidth; and based on the updated corresponding relationship, query to obtain the updated bandwidths of multiple speed-limiting points to limit the speed of sliced packets to be transmitted on the physical link.
  • the network element at the sending end first updates the attachment relationship information based on the updated speed limit point information of the multiple speed limit points; Then, based on the updated attachment relationship information, the rate of the sliced packets to be transmitted on the physical link is limited.
  • the network element at the sending end needs to perform one-time rate limiting of the sliced packets, and the bandwidth used for rate limiting is a fixed bandwidth. That is, after the staff sets the bandwidth, the bandwidth will not be changed. This results in less flexibility in bandwidth for rate limiting.
  • the network element at the sending end may update the speed limit point information of at least one speed limit point among the multiple speed limit points on the physical link, for example, periodically update or update the speed limit point information in real time.
  • the number of bandwidths used for rate limiting is dynamically variable. The multiple bandwidths are adjusted according to the actual situation, so as to ensure that the multiple bandwidths used for speed limiting can effectively adapt to different speed limiting environments and improve the flexibility of the bandwidths used for speed limiting.
  • the network element at the sending end sends the rate-limited slice packet to the physical link.
  • the network element at the sending end after the network element at the sending end performs multi-level rate limiting, as after S302 or S304, the network element at the sending end outputs the output to the physical link through the physical interface on the network element at the sending end.
  • the sliced packets passing through the physical link are transmitted on the basis of ensuring the maximum payload bandwidth of the sliced packets as much as possible, thereby improving the bandwidth utilization.
  • the network element at the sending end uses the obtained bandwidths of multiple speed limiting points on the physical link to limit the speed of the sliced packets to be transmitted on the physical link.
  • the rate of the network element at the sending end has been limited according to the bandwidth of each rate-limiting point. Therefore, when the sliced packets after the rate limit are actually transmitted on the physical link, the bandwidth requirements of each rate limit point have been met.
  • the network element at the sending end directly obtains the bandwidth of the speed-limiting point itself, and performs speed-limiting based on the obtained bandwidth. There is no need to manually determine the speed-limiting bandwidth of each speed-limiting point, which reduces labor costs, simplifies the process of obtaining the bandwidth for speed-limiting, and improves the obtaining efficiency of the bandwidth for speed-limiting.
  • the communication system also supports a bandwidth prediction function, and flexible speed limit can be implemented by predicting changes in bandwidth in advance.
  • the bandwidth prediction function includes the following steps:
  • the network element at the sending end receives the bandwidth prediction information sent by the prediction device.
  • the prediction device corresponds to at least one physical link among the multiple physical links, and the prediction device is used to predict the changed bandwidth of the speed limit point monitored on the corresponding physical link.
  • the setting method of the prediction device may be the same as or similar to the setting method of the monitoring device in the aforementioned S301.
  • each physical link in the aggregated link includes at least one hardware device, and the at least one hardware device includes the aforementioned prediction device.
  • the hardware device where the speed limit point in the at least one hardware device is located is a prediction device, and the speed limit point monitored by any prediction device on the corresponding physical link includes the speed limit point located on the any prediction device. The speed limit point and/or the speed limit point connected to any of the predicted devices.
  • the specified hardware device in the at least one hardware device is a prediction device
  • the speed limit point monitored by the prediction device on the corresponding physical link includes one or more speed limit points on the corresponding physical link Speed limit point.
  • the rate limit point monitored by the prediction device is a rate limit point with variable bandwidth on the corresponding physical link, such as an air interface.
  • the prediction device and the aforementioned monitoring device may be the same device or different devices. Predict the bandwidth change conditions of the monitored speed limit points that are preconfigured in the device. After monitoring that a certain speed limit point satisfies the bandwidth change condition, the prediction device predicts the changed bandwidth of the certain speed limit point, and sends bandwidth prediction information including the predicted bandwidth to the network element at the sending end.
  • the predicted bandwidth is the changed bandwidth of the certain speed limit point predicted by the prediction device after determining that a certain speed limit point monitored on the corresponding physical link satisfies the bandwidth change condition.
  • the bandwidth change condition is: the environment where the speed limit point is located starts to produce a preset change, or the environment where the speed limit point is located produces a preset change after a preset time period; or the bandwidth of the speed limit point starts to change after a preset time period reset etc.
  • the preset change includes: a temperature change with a change range greater than the preset temperature difference, and/or a humidity change with a change range greater than the preset humidity difference, and the like.
  • the prediction device is used to determine whether the speed limit point satisfies the bandwidth change condition through autonomous monitoring; in another optional way, the prediction device is used to receive a trigger message sent by other devices, the trigger The message is used to indicate whether the speed limit point monitored by the prediction device satisfies the bandwidth change condition.
  • the prediction device includes an environment monitoring module, and the prediction device monitors the environment of the speed limit point through the environment monitoring module to determine whether the speed limit point satisfies the bandwidth change condition; for another example, the prediction device is provided with a communication module, and the communication module is used for receiving Set the trigger message sent by the environmental monitoring device near the speed limit point.
  • the prediction device can predict the changed bandwidth of the certain speed limit point in various ways. For example, if the bandwidth change condition is: the environment where the speed limit point is located begins to undergo a preset change, or, the environment where the speed limit point is located generates a preset change after a preset period of time; ) and the change of the bandwidth to determine the changed bandwidth. If the bandwidth change condition is: the bandwidth of the speed limit point is reset after a preset time period, the prediction device obtains the reset bandwidth in advance as the predicted bandwidth.
  • the prediction device may repeatedly send the bandwidth prediction information to ensure that the network element at the sending end can receive the bandwidth prediction information and improve the reliability of reporting the bandwidth prediction information.
  • the network element at the sending end limits the speed of the sliced packets to be passed through a certain speed-limiting point among the sliced packets transmitted on the physical link.
  • the sending end network element may refer to the speed limiting process in the foregoing S302 for the process of limiting the speed of the segmented packets to be passed through a certain speed-limiting point among the segmented packets transmitted on the physical link.
  • the network element at the sending end establishes a corresponding relationship between the identification of the speed-limiting point and the bandwidth.
  • the network element at the sending end firstly updates the corresponding relationship between the identification of the speed limit point and the bandwidth based on the speed limit bandwidth; Slice packets for rate limiting.
  • the bandwidths of the multiple speed limit points obtained by the query include the predicted bandwidth
  • the slice packets to be transmitted on the physical link include the slice packets to pass through the certain speed limit point.
  • the network element at the sending end first updates the attachment relationship information based on the speed-limited bandwidth; and then based on the updated attachment relationship information Limit the rate of sliced packets to be transmitted on the physical link.
  • the attachment relationship information includes the predicted bandwidth, and the slice packets to be transmitted on the physical link include the slice packets to pass through the certain speed limit point.
  • the prediction device predicts the bandwidth that will take effect in advance, and notifies the network element at the sending end.
  • the network element at the sending end responds and adjusts the bandwidth in the network element in advance, thereby realizing lossless bandwidth switching.
  • it can effectively prevent the loss of sliced packets caused by the failure of the rate limit, thereby avoiding the phenomenon of random packet loss.
  • the actual payload bandwidth of the longer sliced packets through the member links (that is, the sliced packets in the unit time)
  • the actual payload passing through a speed limit point) is far less than its maximum payload bandwidth.
  • Such a long-length sliced packet (that is, a sliced packet with a length greater than the minimum length) has a relatively low proportion of the payload to the length of the sliced packet within a unit time (this ratio is called the passing rate), so that the length of the packet is relatively low.
  • the communication loss of the long sliced packet on the member link is relatively large (for example, more than 30%).
  • the network element at the transmitting end directly uses the bandwidth of each rate limiting point and the number of overhead bits to limit the rate of sliced packets of different lengths. Therefore, for sliced packets of different lengths, it can be ensured that they can pass through the maximum payload bandwidth or close to the maximum payload bandwidth as much as possible, thereby reducing communication loss.
  • the network element at the sending end reduces congestion by monitoring traffic, thereby avoiding random packet loss.
  • the network element at the sending end of the embodiment of the present application also supports a flow control back pressure method.
  • Flow control back pressure refers to the flow control of sliced packets through feedback.
  • the process of the flow control back pressure includes:
  • the sender network reduces the bandwidth corresponding to the sum of the length of the sliced packets to pass through the rate-limiting point and the number of overhead bits at the rate-limiting point by discarding service packets.
  • This service packet is used for slicing to obtain sliced packets.
  • the bandwidth corresponding to the sum of the length of the sliced packet at the speed limit and the number of overhead bits at the speed limit is calculated based on a preset conversion algorithm.
  • the units of h1 and h2 are the same, or the units are the same after conversion.
  • the entrances of the connection networks of the multiple speed limiters are connected to the slice module.
  • the first speed limiter is any speed limiter among the plurality of speed limiters, and the first speed limiter point is a speed limit point corresponding to the first speed limiter.
  • the first rate limiter is further configured to detect whether the bandwidth corresponding to the sum of the length of the sliced packet and the number of overhead bits of the first rate limiting point is greater than the bandwidth of the first rate limiting point.
  • the first rate limiter When it is detected that the bandwidth corresponding to the sum of the length of the sliced packet and the number of overhead bits of the first rate limit point is greater than the bandwidth of the first rate limit point, in the first optional manner, the first rate limiter sends the first rate limiter to the first rate limiter.
  • the upstream of the speed limiter feeds back the back pressure information (if the first speed limiter is the entrance of the connection network of multiple speed limiters, the first speed limiter feeds back the back pressure information to the slicing module; if the first speed limiter is not multiple speed limiters are connected to the network, the first speed limiter feeds back pressure information to the upper-level speed limiter), the back pressure information is used to indicate the length of the sliced packet and the overhead of the first speed limit point The bandwidth corresponding to the sum of the number of bits is greater than the bandwidth of the first speed limit point.
  • the back pressure information is fed back by each speed limiter in the reverse order of the order of the speed limit point identification in the attachment relationship information, until it reaches the slicing module; the slicing module discards the service packets to be fragmented, thereby reducing the unit size
  • the number of sliced packets sent within the duration that is, to reduce the packet traffic, thereby reducing the bandwidth corresponding to the sum of the length of the sliced packets reaching the first rate limiter and the number of overhead bits at the rate-limiting point.
  • the first speed limiter is the speed limiter 4, it sends back pressure information to the upper-level speed limiter of the speed limiter 4: the speed limiter 3, and the speed limiter 3 sends the speed limiter to the speed limiter 3.
  • the multiple speed limiters are also connected to the slicing module respectively.
  • the first rate limiter feeds back the back pressure information to the slicing module.
  • the delay from the first rate limiter to the slicing module is small, which can realize rapid adjustment of packet traffic.
  • the connection relationship within the network element at the sending end is relatively complex, and the management cost is high.
  • the network element at the sending end may determine the number of discarded service packets according to the congestion situation of the first rate limiter.
  • the backpressure information includes a congestion level that reflects the severity of the congestion situation.
  • the first rate limiter can make the difference between the bandwidth corresponding to the sum of the length of the sliced packet and the number of overhead bits of the first rate limit point and the bandwidth of the first rate limit point, and use the obtained target difference to query the preset congestion. Correspondence between levels and difference ranges.
  • the congestion level corresponding to the difference range where the target difference is located is carried in the back pressure information.
  • the network element at the sending end queries the correspondence between the congestion level and the number of discards, obtains the target number of discards, and discards the target number of service packets to be discarded. For example, a service packet carries a priority, and the network element at the sending end discards the service packet in the order of priority from low to high. In this way, on the premise of reducing congestion, the service packet with high priority is minimized. throw away.
  • the network element at the sending end Since the network element at the sending end performs flow control by directly discarding the service packets, it avoids random packet loss caused by the loss of sliced packets, especially reduces the packet loss of high-priority service packets, thus ensuring the communication process. reliability.
  • the network element at the sending end adopts the method of flow control back pressure, so that sliced packets of various lengths can be transmitted as much as possible, which effectively improves the transmission efficiency of sliced packets.
  • the network element at the sending end and the network element at the receiving end in the communication system are relative.
  • the network element may be both a sending end network element and a receiving end network element.
  • FIG. 7 is a schematic structural diagram of a communication apparatus 40 provided by an embodiment of the present application.
  • the apparatus 40 is applied to the network element at the sending end, and the apparatus 40 includes a first obtaining module 401 and a first rate limiting module 402 .
  • the first obtaining module 401 is configured to obtain speed limit point information of multiple speed limit points on the physical link for each physical link of the multiple physical links bound by the PLA.
  • the speed limit point information includes bandwidth.
  • the first rate limiting module 402 is configured to limit the rate of sliced packets to be transmitted on the physical link based on the bandwidth of multiple rate limiting points on the physical link.
  • the first acquisition module 401 is configured to receive monitoring information sent by a monitoring device, where the monitoring device corresponds to at least one physical link in a plurality of physical links, and the monitoring information includes the speed limit monitored by the monitoring device on the corresponding physical link speed limit point information.
  • each physical link in the multiple physical links includes at least one hardware device, the hardware device where the speed limit point is located in the at least one hardware device is a monitoring device, and any monitoring device monitors the corresponding physical link.
  • the speed limit point includes a speed limit point located on any monitoring device and/or a speed limit point connected to any monitoring device; the first acquisition module 401 is used for: receiving monitoring data sent by the monitoring device on each physical link information.
  • the first obtaining module 401 is configured to: receive a bandwidth setting instruction, where the bandwidth setting instruction includes speed limit point information of one or more speed limit points in multiple physical links.
  • the apparatus further includes: a second obtaining module 403 .
  • the second obtaining module 403 is configured to: obtain the speed limit point information of any speed limit point when any speed limit point information among the plurality of speed limit points changes; and/or periodically obtain multiple speed limit points speed limit information.
  • the apparatus further includes: a receiving module 404 and a second speed limiting module 405 .
  • the receiving module 404 is configured to receive bandwidth prediction information sent by the prediction device.
  • the prediction device corresponds to at least one physical link of the plurality of physical links.
  • the bandwidth prediction information includes predicted bandwidth.
  • the predicted bandwidth is the bandwidth after the bandwidth of any speed limit point on the physical link corresponding to the prediction device predicted by the prediction device is changed.
  • the second rate limiting module 405 is configured to, based on the predicted bandwidth, limit the rate of the sliced packets to be transmitted through any rate-limiting point among the sliced packets to be transmitted on the physical link.
  • the first speed limiting module 402 in the communication device 40 may perform speed limiting based on the connection relationship information.
  • the first rate limiting module 402 is configured to: acquire attachment relationship information, where the attachment relationship information is used to describe the connection relationship of each speed limit point on the physical link.
  • the attachment relationship information is established based on the identification of each speed limit point on the physical link and the speed limit point type of each speed limit point. Based on the bandwidth of multiple rate-limiting points on the physical link, and according to the connection relationship of each rate-limiting point in the attachment relationship information, multi-level rate-limiting is performed on the sliced packets to be transmitted on the physical link.
  • the apparatus 40 further includes: a processing module 406 for each speed limit point recorded in the attachment relationship information.
  • a processing module 406 for each speed limit point recorded in the attachment relationship information.
  • the physical link includes m member links, where m is a positive integer.
  • the speed limit point information further includes: the identification of the speed limit point and the speed limit point type.
  • the speed limit point type includes main road speed limit point or branch road speed limit point.
  • the speed limit point of the main road is the speed limit point jointly passed by the m member links
  • the speed limit point of the branch road is the speed limit point passed by the m member links respectively.
  • the information included in the attachment relationship information of the physical link refers to the description shown in FIG. 4 , that is, the first branch and the second branch, and details are not described herein again.
  • the identification of the speed limit point on the branch satisfies at least one of the following: when the bandwidth of the two speed limit points is the same and the number of overhead bits is different, the two The identifiers of the speed limit points are arranged in descending order of the corresponding overhead bits in the branch; when the number of overhead bits of the two speed limit points is the same and the bandwidth is different, the identifiers of the two speed limit points are arranged in the branch according to the number of overhead bits. Corresponding bandwidths are arranged in ascending order.
  • the processing module 406 is configured to delete the two speed limit points recorded in the branch when the number of overhead bits of the two speed limit points is the same and the bandwidth is different.
  • the identification of the speed limit point with the larger bandwidth is the updated branch; or, when the number of overhead bits of the two speed limit points is different and the bandwidth is the same, delete the record in the branch with the higher number of overhead bits of the two speed limit points.
  • the identification of the smaller speed limit point is obtained from the updated branch; or, when the number of overhead bits of the two speed limit points is the same and the bandwidth is the same, the identification of one speed limit point of the two speed limit points recorded in the branch is deleted. Get the updated branch.
  • FIG. 8 is a schematic structural diagram of a communication apparatus 50 according to another embodiment of the present application.
  • the communication device 50 includes a processing chip 501
  • the processing chip 501 includes a processing circuit 5011 and a communication interface 5012 .
  • the processing circuit is used in the communication method performed by the transmitter in the foregoing embodiments of the present application.
  • the processing circuit 5012 may be a processing chip or a field programmable gate array (Field Programmable Gate Array, FPGA).
  • the processing chip may be an integrated circuit (Application Specific Integrated Circuit, ASIC) chip; the communication interface 5012 is used for the processing circuit 5011 to communicate with other devices.
  • ASIC Application Specific Integrated Circuit
  • the communication interface 5012 is used for the processing circuit 5011 to communicate with the network element at the receiving end through the aggregated link.
  • the communication interface 5012 includes an input interface and an output interface.
  • the communication interface 5012 may be any one or any combination of the following: a network interface (eg, an Ethernet interface), a wireless network card, or the like.
  • the processing circuit 5012 includes a cache structure, such as a storage structure inside an FPGA or ASIC chip, for caching the speed limit point information.
  • the communication apparatus 501 further includes: a memory for buffering the speed limit point information.
  • the memory is flash memory.
  • An embodiment of the present application provides a communication system, where the communication system includes a sending end network element, a receiving end network element, and a plurality of physical links bound by PLA, and the sending end network element includes the communication device provided by the embodiment of the present application, Such as the communication device 40 or the communication device 50 .
  • the structure of the communication system may refer to FIG. 1 or FIG. 2 .
  • the terms “first”, “second” and “third” are used for descriptive purposes only and should not be understood as indicating or implying relative importance.
  • the term “at least one” refers to one or more, and the term “plurality” refers to two or more, unless expressly limited otherwise.
  • A refers to B, which means that A is the same as B or A is a simple variation of B.
  • the communication device provided in the above embodiment executes the communication method
  • only the division of the above functional modules is used as an example.
  • the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
  • the communication apparatus and the communication method embodiments provided by the above embodiments belong to the same concept, and the specific implementation process thereof is detailed in the method embodiments, which will not be repeated here.

Abstract

本申请公开了一种通信方法、装置及系统，属于通信领域，用于业务报文的传输，应用于发送端网元。所述方法包括：对于通过物理链路聚合PLA捆绑的多条物理链路中每条物理链路，获取所述物理链路上的多个限速点的限速点信息，所述限速点信息包括带宽；基于所述物理链路上的多个限速点的带宽，对待在所述物理链路上传输的切片报文进行限速。本申请能够降低成本，简化带宽获取流程。

Description

通信方法、装置及系统
本申请要求于2020年11月27日提交的、申请号为202011359748.2、申请名称为“通信方法、装置及系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，特别涉及一种通信方法、装置及系统。
背景技术
物理链路聚合(Physical Link Aggregation，PLA)技术，也称主干技术(Trunking)或捆绑技术(Bonding)，指的是将多条物理链路进行捆绑，从而提供一个逻辑上较大的管道来进行业务报文传输的技术。
基于PLA技术的通信系统包括聚合链路以及通过该聚合链路连接的发送端网元(Network Element，NE)和接收端网元。聚合链路包括多条物理链路的每一条上配置有一个或多条成员链路。在该通信系统中，发送端网元用于对业务报文进行切片得到多个切片报文，然后将多个切片报文分发到聚合链路中的多条成员链路上。接收端网元用于接收多个切片报文并排序，然后将排序后的切片报文进行重组得到业务报文。若任一切片报文在传输过程中被丢弃，会导致切片报文重组失败，出现随机丢包。
相关技术中，为了避免传输过程中随机丢包的现象发生，对于每条成员链路，在发送端网元通过一次限速(也称流量整形(traffic shaping))将经过该成员链路的切片报文所占用的带宽设置为小于或等于该成员链路的带宽。该成员链路的带宽是由工作人员设置的用于限速的带宽。工作人员需要计算成员链路上的各个限速点的限速带宽，并将该各个限速点的限速带宽中的最小值确定为该成员链路的带宽。每个限速点的限速带宽需要基于每个限速点的带宽、开销比特、通过该限速点的切片报文的最小长度以及相邻限速点的带宽等综合确定，人工成本较高。导致用于限速的带宽的获取过程复杂，获取效率较低。
发明内容
本申请实施例提供了一种通信方法、装置及系统，提升限速效率。
第一方面，提供一种通信方法，应用于发送端网元，该方法包括：对于通过PLA捆绑的多条物理链路中每条物理链路，发送端网元获取该物理链路上的多个限速点的限速点信息。其中，该限速点信息包括带宽。发送端网元基于该物理链路上的多个限速点的带宽，对待在该物理链路上传输的切片报文进行限速。
发送端网元通过获取的物理链路上的多个限速点的带宽，对待在该物理链路上传输的切片报文进行限速。如此，切片报文在经过物理链路之前，已经在发送端网元处按照各个限速点的带宽进行了限速。因此限速后的切片报文在实际在物理链路上传输时已经满足了各个限速点的带宽要求。在此基础上，发送端网元直接获取限速点本身的带宽，并基于获取的带宽进行限速。无需人工确定每个限速点的限速带宽，减少人工成本，简化了用于限速的带宽的获取过程，提高了用于限速的带宽的获取效率。
可选地，限速点信息还包括：限速点的开销比特数。示例地，基于该物理链路上的多个限速点的带宽，对待在该物理链路上传输的切片报文进行限速的过程包括：对于该多个限速点的每个限速点，将该限速点的带宽与开销比特数对应的带宽之差作为阈值带宽；控制待经过该限速点的切片报文的单位时间内的通过速度小于该限速点的阈值带宽。其中，开销比特数对应的带宽为该开销比特数与预设的报文传输速率的乘积。
在一种可选方式中，发送端网元还可以更新物理链路上的多个限速点中至少一个限速点的限速点信息，并基于更新后的多个限速点的带宽，对待在物理链路上传输的切片报文进行限速。具体地，有两种更新方式。第一种方式，在该多个限速点中的任一限速点信息变化时，获取该任一限速点的限速点信息；发送端网元在多个限速点中的任一限速点信息变化时，获取该任一限速点的限速点信息。如此可以保证发送端网元对变化的限速点信息的实时获取，保证限速的及时准确。第二种方式，周期性获取该多个限速点的限速点信息。
发送端网元可以通过多种方式获取物理链路上的多个限速点的限速点信息。在第一种实现方式中，发送端网元获取该物理链路上的多个限速点的限速点信息的过程，包括：发送端网元接收监控设备发送的监控信息，该监控设备与该多条物理链路中的至少一条物理链路对应。该监控信息包括该监控设备在对应的物理链路上监控到的限速点的限速点信息。
可选地，该多条物理链路中每条物理链路包括至少一个硬件设备，该至少一个硬件设备中限速点所在硬件设备为该监控设备。任一监控设备在对应的物理链路上监控到的限速点包括位于该任一监控设备上的限速点和/或与该任一监控设备连接的限速点；该接收监控设备发送的监控信息的过程，包括：接收该每条物理链路上的监控设备发送的监控信息。
在第二种实现方式中，发送端网元获取该物理链路上的多个限速点的限速点信息的过程，包括：接收带宽设置指令。该带宽设置指令包括该多条物理链路中一个或多个限速点的限速点信息。
在一种可选示例中，发送端网元具有或连接有输入设备，用户通过该输入设备输入带宽设置指令。相应的，发送端网元接收该带宽设置指令。该带宽设置指令包括用户能够获取的一个或多个限速点的限速点信息。该限速点信息通常为带宽固定的限速点的限速点信息。如此，用户通常仅需设置一次限速点信息，减少用户操作。
前述第一种实现方式和第二种实现方式可以组合执行。例如，发送端网元获取的限速点信息中包括监控设备上报的限速点信息以及通过带宽设置指令设置的限速点信息。
接收端网元获取的限速点信息可能有重复的情况。针对这种情况，发送端网元在接收到限速点信息后，可以对接收的限速点信息进行去重处理，得到去重处理后的限速点信息。示例地，对于同一限速点，若接收到多个限速点信息，获取该多个限速点信息中最新的限速点信息(即设置时间或发送时间距离当前时刻最近的限速点信息)作为去重后的限速点信息，如此可以保证限速点信息的及时性和有效性。
通过限速点信息的冗余上报可以降低限速点信息的漏发风险，提高发送端网元所接收的限速点信息的可靠性。
本申请实施例中，通信系统还支持带宽预测功能。通过提前预测带宽的变化可以实现灵活限速。该带宽预测功能包括以下步骤：接收预测设备发送的带宽预测信息。该预测设备与该多条物理链路中的至少一条物理链路对应。该带宽预测信息包括预测带宽。该预测带宽为该预测设备在对应的物理链路上监控到的某一限速点满足带宽变化条件后，预测的 该某一限速点的变化后的带宽。基于该预测带宽，对待在该物理链路上传输的切片报文中待经过该某一限速点的切片报文进行限速。
需要说明的是，预测设备可以重复发送该带宽预测信息，以保证发送端网元能够收到该带宽预测信息，提高带宽预测信息上报的可靠性。
本申请实施例中，当发送端网元支持带宽预测功能时，通预测设备提前预测即将生效的带宽，并通知发送端网元。由该发送端网元进行响应，提前进行该网元内的带宽调整，从而实现无损的带宽切换。尤其对于一些优先级较高的业务报文，有效防止因为限速失效所导致的切片报文丢失，从而避免随机丢包现象。
可选地，该基于该物理链路上的多个限速点的限速点信息，对待在该物理链路上传输的切片报文进行限速的过程，包括：获取挂接关系信息，该挂接关系信息用于描述该物理链路上各个限速点的连接关系。该挂接关系信息基于该物理链路上的每个限速点的标识以及该每个限速点的限速点类型建立。基于该物理链路上的多个限速点的带宽，按照该挂接关系信息中各个限速点的连接关系，对待在该物理链路上传输的切片报文进行多级限速。
示例地，基于该物理链路上的多个限速点的带宽，按照该挂接关系信息中各个限速点的连接关系，对待在该物理链路上传输的切片报文进行多级限速的过程包括：对于该挂接关系信息中记录的每个限速点，将该限速点的带宽与开销比特数对应的带宽之差作为阈值带宽。控制待经过该限速点的切片报文的单位时间内的通过速度小于该限速点的阈值带宽。
其中，该物理链路包括m条成员链路，m为正整数。该限速点信息还包括：限速点的标识和限速点类型。该限速点类型包括主路限速点或支路限速点。该主路限速点为该m条成员链路共同经过的限速点，该支路限速点为该m条成员链路分别经过的限速点。
假设该物理链路的挂接关系信息包括：与该m条成员链路对应的m个第一分支(例如，m个第一分支与m条成员链路一一对应)，以及与该m个第一分支连接的一个第二分支；则每个第一分支用于记录对应的成员链路上的限速点类型为支路限速点的限速点的标识，该第二分支用于记录该物理链路上的限速点类型为主路限速点的限速点的标识。
在一种可选实现方式中，在物理链路的挂接关系信息中，对于第一分支和第二分支中的每个分支，该分支记录的限速点的标识可以随机排列。
在另一种可选方式中，对于第一分支和第二分支中的每个分支，该分支上的限速点的标识可以按照预设规则排列。例如，该分支上的限速点的标识满足以下至少一种情况：
第一种情况，当两个限速点的带宽相同，且开销比特数不同时，该两个限速点的标识在分支中按照对应的开销比特数由大到小的顺序排列。如此在进行限速时，优先采用前一个限速点的带宽和开销比特数进行限速，使得限速后的切片报文在到达后一个限速点时，已经满足后一个限速点的限速要求，降低切片报文在后一个限速点的通过时延。
第二种情况，当两个限速点的开销比特数相同，且带宽不同时，该两个限速点的标识在分支中按照对应的带宽由小到大的顺序排列。如此在进行限速时，优先采用前一个限速点的带宽和开销比特数进行限速，使得限速后的切片报文在到达后一个限速点时，已经满足后一个限速点的限速要求，降低切片报文在后一个限速点的通过时延。
在本申请实施例中，每个第一分支通常记录对应的成员链路上的所有限速点类型为支路限速点的限速点的标识；第二分支通常记录物理链路上的所有限速点类型为主路限速点的限速点的标识。如此，保证物理链路的挂接关系信息覆盖物理链路上的所有限速点的标 识，避免出现限速点的遗漏，影响限速结果。
在实际实现时，对于第一分支和第二分支中的每个分支，还可以将该分支所记录的限速点的标识进行整合，保留需要整合的限速点中的一个限速点的标识。如此，每个第一分支记录的限速点的标识数量小于或等于对应的成员链路上的限速点类型为支路限速点的限速点的标识总数；第二分支记录的限速点的标识数量小于或等于物理链路上的限速点类型为主路限速点的限速点的标识总数。
例如，对于该第一分支和该第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽不同，发送端网元可以删除该分支中记录的该两个限速点中带宽较大的限速点的标识得到更新后的分支。又例如，对于该第一分支和该第二分支中的每个分支，当两个限速点的开销比特数不同，且带宽相同，发送端网元可以删除该分支中记录的该两个限速点中开销比特数较小的限速点的标识得到更新后的分支。再例如，对于该第一分支和该第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽相同，发送端网元可以删除该分支中记录的该两个限速点中一个限速点的标识得到更新后的分支。
基于前述更新后的分支进行限速可以达到基于更新前的分支进行限速的相同效果，但是更新后的分支记录的限速点的标识数量更少，复杂度更低。
本申请实施例的发送端网元还支持流控反压的方式。流控反压指的是通过反馈的方式实现切片报文的流量控制。本申请实施例中，该流控反压的过程包括：对于该挂接关系信息记录的每个限速点，当待经过该限速点的切片报文的长度和该限速点的开销比特数之和所对应的带宽大于该限速点的带宽。通过丢弃业务报文的方式降低待经过该限速点的切片报文的长度和该限速点的开销比特数之和所对应的带宽，该业务报文用于切片得到切片报文。
第二方面，提供一种通信装置。该通信装置包括：处理芯片，所述处理芯片包括：处理电路及通信接口，该处理电路用于执行第一方面任一的通信方法；该通信接口用于供该处理电路与其他装置进行通信。该处理电路可以为处理芯片或现场可编程逻辑门阵列(Field Programmable Gate Array，FPGA)。
第三方面，提供一种通信装置。该通信装置包括至少一个模块，该至少一个模块可以用于实现上述第一方面或者第一方面的各种可能实现方式提供的通信方法。
第四方面，本申请提供一种通信系统。所述通信系统包括发送端网元、接收端网元和以及通过物理链路聚合PLA捆绑的多条物理链路，该发送端网元包括第二方面或第三方面所述的通信装置。
第五方面，本申请提供一种计算机设备，该计算机设备包括处理器和存储器。该存储器存储计算机指令；该处理器执行该存储器存储的计算机指令，使得该计算机设备执行上述第一方面或者第一方面的各种可能实现提供的方法。
第六方面，本申请提供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计算机指令，该计算机指令指示计算机设备执行上述第一方面或者其各种可能实现提供的方法。
第七方面，本申请提供一种计算机程序产品，该计算机程序产品包括计算机指令，该计算机指令存储在计算机可读存储介质中。计算机设备的处理器可以从计算机可读存储介质读取该计算机指令，处理器执行该计算机指令，使得该计算机设备执行上述第一方面其 各种可能实现提供的方法。
第八方面，提供一种芯片。该芯片可以包括可编程逻辑电路，当该芯片运行时用于实现如第一方面任一的通信方法。
综上所述，本申请实施例中，发送端网元通过获取的物理链路上的多个限速点的带宽，对待在该物理链路上传输的切片报文进行限速。如此，切片报文在经过物理链路之前，已经在发送端网元处按照各个限速点的带宽进行了限速。因此限速后的切片报文在实际在物理链路上传输时已经满足了各个限速点的带宽要求。在此基础上，发送端网元直接获取限速点本身的带宽，并基于获取的带宽进行限速。无需人工确定每个限速点的限速带宽，减少人工成本，简化了用于限速的带宽的获取过程，提高了用于限速的带宽的获取效率。
附图说明
图1是本申请提供的一种通信方法所涉及的通信系统的结构示意图；
图2是本申请提供的一种通信方法所涉及的通信系统的通信原理示意图；
图3是本申请实施例提供的一种通信方法的流程示意图；
图4是本申请实施例提供的一种示意性的挂接关系示意图；
图5是图1所示的物理链路的挂接关系信息示意图；
图6是图5所示的物理链路的挂接关系信息所对应的限速器的连接网络示意图；
图7是本申请实施例提供的一种通信装置的结构示意图；
图8是本申请实施例提供的另一种通信装置的结构示意图。
具体实施方式
为使本申请的原理和技术方案更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。
图1是本申请实施例提供的一种通信方法所涉及的通信系统的结构示意图。如图1所示，该通信系统包括：发送端网元101、接收端网元102和聚合链路。该聚合链路包括捆绑的多条物理链路。图1假设聚合链路包括2条物理链路，分别是物理链路1031和物理链路1032。在发送端网元处，一个聚合链路对应一个逻辑接口，该逻辑接口也称为链路聚合接口。组成链路聚合接口的每个物理接口(也称物理端口或硬件接口)所对应的链路称为物理链路。该物理链路可以指连接两个网元的直接路径(即两个网元之间不存在中间网元)或者间接路径(即两个网元之间存在中间网元)。该物理链路通常为实体链路。如此，发送端网元101和接收端网元102之间的一条物理链路为发送端网元101上的一个物理接口到接收端网元102上的一个物理接口之间的直接或间接路径。发送端网元101用于对业务报文进行切片得到多个切片报文。聚合链路用于传输该多个切片报文。接收端网元102用于接收多个切片报文并排序，然后将排序后的切片报文进行重组得到业务报文。示例地，网元可以是路由器、交换机、或其他类型的网元。
在本申请实施例中，该聚合链路包括一种或多种类型的物理链路。例如，该聚合链路包括：以太链路、微波链路、光传送网(Optical Transport Network，OTN)链路和改进型OTN链路(也称增强型OTN链路或者演进型OTN链路)的至少一种类型的物理链路。该聚合链路中的物理链路可以由同一运营商部署，也可以由不同运营商部署。
以聚合链路包括微波链路为例。采用微波链路的PLA技术可以分为普通PLA(也称传统PLA)技术、增强型PLA(Enhanced Physical Link Aggregation，EPLA)技术和超级双频(Super Dual Band，SDB)技术等几种类型。其中，在普通PLA技术中，聚合链路包括两条微波链路。在增强型PLA技术中，聚合链路包括至少三条微波链路；在SDB技术中，聚合链路包括两种不同频段的微波链路。例如，包括常规频段的微波链路以及E-BAND(E-频段)的微波链路。该常规频段的频率为7-38吉赫(GHz)，该E-BAND的频率为71-86GHz。
图2是本申请实施例提供的一种通信方法所涉及的通信系统的通信原理示意图。如图2所示，每条物理链路上配置有一个或多条成员链路(也可视为将物理链路划分为一个或多条成员链路)，则一个聚合链路包括多条成员链路，聚合链路的成员链路的数量大于或等于物理链路的数量。其中，属于同一物理链路的成员链路共同组成链路聚合接口的一个物理接口，属于同一物理链路的不同成员链路分别使用不同的通信接口，属于同一物理链路的至少一条成员链路与至少一个通信接口一一对应。该通信接口可以为空口或有线口，该有线口可以为光接口或电接口。当一条成员链路使用空口通信时，该成员链路包括该空口连接的无线链路；当一条成员链路使用有线口进行通信时，该成员链路包括该有线口连接的通信线缆。
在图2所示的通信系统中，发送端网元用于对业务报文执行报文切片流程，以得到多个切片报文，再对多个切片报文执行报文分发流程，以将多个切片报文分发到聚合链路中的多条成员链路上；接收端网元用于执行切片收集过程，以将成员链路上传输的多个切片报文进行收集；然后对收集的多个切片报文执行切片缓存流程，以将该多个切片报文进行排序；之后再执行报文重组流程，以将排序后的切片报文重组得到业务报文。
例如，图2中假设聚合链路包括n条成员链路，分别为成员链路1至成员链路n，n≥2。发送端网元101对业务报文执行报文切片流程以及报文分发流程后，将切片报文1、3、7和8分发至成员链路1上，将切片报文2、4、5和6分发至成员链路n上；接收端网元102通过执行切片收集流程、切片缓存流程和报文重组流程，得到切片报文1-8组成的业务报文。
若任一切片报文在传输过程中被丢弃，会导致接收端网元的切片报文重组失败，出现发送端网元和接收端网元之间的随机丢包。因此，需要保证成员链路上的每个限速点都能允许切片报文通过。限速点是相对于发送端网元所发送的切片报文而言的，其是对发送端网元发送的切片报文产生限速作用的点。例如，限速点可以是背板、单板或设备上的物理接口。该物理接口可以为连接器、光纤接头、单板插头或以太接口等。单板可以包括中频板。限速点还可以是通信接口，如空口或有线口。发送端网元发送的切片报文在经过限速点时，限速点通常会在切片报文上封装额外的开销比特(或者开销字节)。限速点可以有带宽可变的限速点和带宽固定的限速点两种类型。该带宽可变的限速点的带宽容易受到外界环境，如天气的影响。示例地，该限速点为空口(例如微波空口)。该带宽固定的限速点的带宽不受外界环境的影响，在通信系统中，该限速点的带宽在聚合链路部署之后不再更改。例如，一个以太网的物理端口的带宽为1Gpbs(千兆比特每秒)。
为了便于读者对聚合链路中的限速点以及成员链路的进一步理解，下面以图1为例进行说明。对于物理链路1031，发送端网元101生成的切片报文依次经过发送端网元101的背板1、单板1、单板2以及接收端网元102的背板2，从而传输至接收端网元102。其中，该背板1集成在发送端网元101上，其可以与发送端网元101为一体结构。该背板2集成 在接收端网元102上，其可以与接收端网元102为一体结构。背板1具有连接器A、单板1具有用于与背板连接的连接器B以及用于与接收端网元102通信的空口C1和C2。连接器A、连接器B、空口C1和空口C2分别对发送端网元101发送的切片报文产生限速作用。则物理链路1031的限速点包括：连接器A、连接器B、空口C1和空口C2。物理链路1031包括两条成员链路，分别是经过连接器A、连接器B和空口C1的成员链路，以及经过连接器A、连接器B和空口C2的成员链路。对于物理链路1032，发送端网元生成的切片报文依次经过发送端网元101、从设备1和从设备2传输至接收端网元102。发送端网元101具有以太接口D、从设备1具有用于与发送端网元101连接的以太接口E以及用于与接收端网元102通信的空口F。以太接口D、以太接口E、空口F分别对发送端网元101发送的切片报文产生限速作用。则物理链路1032的限速点包括：以太接口D、以太接口E、空口F，物理链路1032包括一条成员链路，即经过以太接口D、以太接口E、空口F的成员链路。
需要说明的是，在同一成员链路上带宽和开销比特数均相同的两个限速点的限速效果相同(即切片报文经过该两个限速点后的限速效果和经过该两个限速点中任意一个后的限速效果是相同的)，两者可以视为同一限速点。例如，在连接器A和连接器B连接(如插接)后，该两个连接器对发送端网元101发送的切片报文的限速作用相同(即切片报文经过该两个连接器后的限速效果和经过该两个连接器中任意一个后的限速效果是相同的)。因此，两者可以视为同一连接器，在物理链路上可以视为同一限速点。同理，以太接口D和以太接口E通过以太口连线连接，由于以太口连线通常不具有限速作用，因此，该两个以太接口对发送端网元101发送的切片报文的限速作用相同(即切片报文经过以太接口D、以太口连线和以太接口E后的限速效果和经过该以太接口D和以太接口E中任意一个后的限速效果是相同的)。如此，两个以太接口可以视为同一以太接口，在物理链路上可以视为同一限速点。将物理链路上限速作用相同的限速点(例如相邻的且限速作用相同的限速点)作为一个限速点进行处理，可以减少获取的物理链路的复杂度，提高对物理链路的管理效率。
图1中的背板2具有连接器，单板2具有连接器和空口，接收端网元102具有以太接口，从设备2具有以太接口和空口。图1假设这些物理接口和通信接口并未对发送端网元发出的切片报文产生限速作用，也即是对于切片报文来说，不是限速点。因此图1未示出这些物理接口和通信接口。但实际实现时，物理链路上，无论是发送端网元所在侧还是接收端网元所在侧都可能存在限速点，本申请实施例对此不做限定。
在通信系统中，每个限速点允许通过的切片报文的带宽受到以下几个因素的影响：
1、限速点添加的开销比特数：每个切片报文在传输过程中，每个限速点都会在该切片报文上添加一些开销比特，不同的限速点，添加的开销比特数不同。相同的限速点，添加的开销比特数相同。
2、限速点的带宽。
3、切片报文的长度：对于不同切片场景，切片报文的长度可能不同，例如切片报文的长度可以为64字节或1518字节等。对于不同长度的报文，由于限速点需要为切片报文添加开销比特，如此导致单位时长内通过同一限速点的不同长度的报文中携带的净荷(payload)不同。该净荷指的是报文实际携带的数据。其中，切片报文在单位时间通过一个限速点的净荷称之为最大净荷带宽，该最大净荷带宽与切片报文的长度正相关。也即是，切片报文的长度越长，其最大净荷带宽越大，切片报文传输的收益越高。
相关技术中，考虑到以上三点因素，并且为了保证不同长度的切片报文均能通过限速点，通常考虑最恶劣的传输情况来确定成员链路的用于限速的带宽。相关技术中采用如下方式确定成员链路的带宽：对于每条成员链路，工作人员基于该成员链路上各个限速点的带宽、开销比特、通过该限速点的切片报文的最小长度以及相邻限速点的带宽等确定各个限速点的限速带宽；再将各个限速点的限速带宽中的最小值确定为该成员链路的带宽。获取各个限速点的限速带宽的过程复杂，人工成本较高。如此导致用于限速的带宽的获取过程复杂，获取效率较低。
图3是本申请实施例提供的一种通信方法的流程示意图。该通信方法应用于图1或图2所示的通信系统。该方法可以降低人工成本，简化用于降速的带宽的获取过程，提高用于降速的带宽的获取效率。本申请实施例提供的通信方法中，对于通过PLA捆绑的多条物理链路(即聚合链路)中每条物理链路，发送端网元执行的处理动作相同。下面以一条物理链路为例进行说明，针对其他物理链路的处理动作参考针对该一条物理链路的处理动作。如图3所示，该方法包括如下步骤。
S301、发送端网元获取物理链路上的多个限速点的限速点信息，该限速点信息包括带宽(即实际的带宽)。
在本申请实施例中，发送端网元可以通过多种方式获取物理链路上的多个限速点的限速点信息。在第一种实现方式中，发送端网元接收监控设备上报的限速点信息。该监控设备与聚合链路中的至少一条物理链路对应，该监控设备用于监控对应的物理链路上限速点的限速点信息。则该发送端网元接收监控设备上报的限速点信息的过程包括：监控设备向发送端网元发送监控信息，该监控信息包括监控设备在对应的物理链路上监控到的限速点的限速点信息；相应的，发送端设备接收监控设备发送的监控信息。可选地，该监控信息还包括信息标识。该信息标识用于唯一标识监控信息，其用于与发送端网元中传输的其他控制信息进行区分。该监控信息还可以包括优先级，监控设备可以按照该监控信息的优先级确定发送监控信息的先后顺序。示例地，该监控信息的优先级为通信系统的最高优先级。
在一种可选示例中，聚合链路中每条物理链路包括至少一个硬件设备，该至少一个硬件设备包括前述监控设备，监控设备对应的物理链路为该监控设备所在的物理链路。每条物理链路上的监控设备向发送端网元发送监控信息；相应的，发送端网元接收该每条物理链路上的监控设备发送的监控信息。示例地，该硬件设备为能够向发送端网元发送信息的设备，其可以为背板、单板、从设备或其他结构。本申请实施例以以下两种情况为例进行说明：
在第一种情况下，该至少一个硬件设备中的限速点所在硬件设备为监控设备，任一监控设备在对应的物理链路上监控到的限速点包括位于该任一监控设备上的限速点和/或与该任一监控设备连接的限速点。
以图1为例，背板1和单板1可以均为物理链路1031上的监控设备，从设备1为物理链路1032上的监控设备。其中，当任一监控设备在对应的物理链路上监控到的限速点包括位于该任一监控设备上的限速点时，背板1发送的监控信息包括：限速点A的限速点信息；单板1发送的监控信息包括：限速点B、C1和C2的限速点信息；从设备1发送的监控信息包括：限速点E和F的限速点信息。当任一监控设备在对应的物理链路上监控到的限速点包括与任一监控设备连接的限速点时，背板1发送的监控信息可以包括：限速点B、C1 和C2的限速点信息；单板1发送的监控信息包括：限速点A的限速点信息；从设备1发送的监控信息包括：限速点D的限速点信息。当任一监控设备在对应的物理链路上监控到的限速点包括位于该任一监控设备上的限速点以及与该任一监控设备连接的限速点时，背板1发送的监控信息包括：限速点A、B、C1和C2的限速点信息；单板1发送的监控信息包括：限速点A、B、C1和C2的限速点信息；从设备1发送的监控信息包括：限速点D、E和F的限速点信息。
在第二种情况下，该至少一个硬件设备中的指定硬件设备为监控设备，该监控设备在对应的物理链路上监控到的限速点包括该对应的物理链路上的一个或多个限速点，例如该物理链路上的所有限速点。
以图1为例，假设背板1、单板1、单板2和背板2中的指定硬件设备为单板1；从设备1和从设备2中的指定硬件设备为从设备1，则单板1为物理链路1031上的监控设备，从设备1为物理链路1032上的监控设备。其中，单板1发送的监控信息包括：限速点A、B、C1和C2的限速点信息；从设备1发送的监控信息包括：限速点D、E和F的限速点信息。
值得说明的是，一条物理链路包括发送端链路，该发送端链路为沿远离发送端网元的方向确定的首个限速点(如物理接口)到最后一个限速点(如空口)之间的链路，由于该发送端链路与发送端网元较近，而该发送端链路之外的硬件设备距离发送端网元较远，因此，可以将指定硬件设备(即监控设备)设置在发送端链路上。当该指定硬件设备为发送端链路上的硬件设备时，可以实现监控信息的快速上报，减少监控信息的传输时延。
前述第一种可选示例中，监控设备是具有通信功能的结构，其能够向发送端网元发送信息。由于传统的背板不具有通信功能，因此，若需要将背板配置为监控设备，该背板为具有通信功能的智能背板。
在另一种可选示例中，监控设备位于聚合链路之外，用于监控聚合链路中的一个或多条物理链路上的限速点，监控设备从所监控的物理链路上的硬件设备上采集各个限速点的限速点信息，并将采集的限速点信息发送给发送端设备。
在第二种实现方式中，发送端网元通过人工配置的方式获取限速点信息。则发送端网元接收监控设备上报的限速点信息的过程包括：接收带宽设置指令，该带宽设置指令包括聚合链路的多条物理链路中一个或多个限速点的限速点信息。
在一种可选示例中，发送端网元具有或连接有输入设备，用户通过该输入设备输入带宽设置指令，相应的，发送端网元接收该带宽设置指令。该带宽设置指令包括用户能够获取的一个或多个限速点的限速点信息。该限速点信息通常为带宽固定的限速点的限速点信息。如此，用户通常仅需设置一次限速点信息，减少用户操作。
值得说明的是，物理链路上的首个限速点通常是发送端网元的物理接口，如图1所示的发送端网元上的背板1的连接器A，以及发送端网元的以太接口D。因此，在实际实现时，除了接收监控设备上报的该首个限速点的限速点信息，或者接收人工配置的该首个限速点的限速点信息之外，发送端网元还可以直接获取自身的物理接口的限速点信息。
本申请实施例在实际实现时，前述第一种实现方式和第二种实现方式可以组合执行。例如，发送端网元获取的限速点信息中包括监控设备上报的限速点信息以及通过带宽设置指令设置的限速点信息。
发送端网元获取的限速点信息可能有重复的情况。例如，在前述第一种实现方式的第一种情况下，当任一监控设备在对应的物理链路上监控到的限速点包括位于任一监控设备上的限速点以及与任一监控设备连接的限速点时，位于同一物理链路上的监控设备所监控的限速点可能存在重叠，从而使得上报的同一限速点的限速点信息有重复。在前述第一种实现方式和第二种实现方式结合的情况下，对于同一限速点，可能存在监控设备上报的限速点信息和人工设置的限速点信息重复的情况。针对这种情况，发送端网元在接收到限速点信息后，可以对接收的限速点信息进行去重处理，得到去重处理后的限速点信息。示例地，对于同一限速点，若接收到多个限速点信息，获取该多个限速点信息中最新的限速点信息(即设置时间或发送时间距离当前时刻最近的限速点信息)作为去重后的限速点信息，如此可以保证限速点信息的及时性和有效性。
通过限速点信息的冗余上报可以降低限速点信息的漏发风险，提高发送端网元所接收的限速点信息的可靠性。
S302、发送端网元基于物理链路上的多个限速点的带宽，对物理链路上传输的切片报文进行限速。
如前所述，由于每个限速点允许通过的切片报文还受限速点添加的开销比特数的影响，因此，限速过程还需要考虑限速点添加的开销比特数。则对于该多个限速点的每个限速点，发送端网元可以将该限速点的带宽与开销比特数对应的带宽之差作为阈值带宽，控制待经过该限速点的切片报文的单位时间内的通过速度小于该限速点的阈值带宽。其中，开销比特数对应的带宽为该开销比特数与预设的报文传输速率的乘积。
在本申请实施例中，发送端网元可以先确定物理链路上各个限速点的连接关系，基于此再进行限速。则基于物理链路上的多个限速点的限速点信息，对物理链路上传输的切片报文进行限速的过程可以包括：
B1、发送端网元获取挂接关系信息，该挂接关系信息用于描述物理链路上各个限速点的连接关系。
发送端网元在获取了多个限速点的限速点信息后，基于获取的限速点信息建立挂接关系图。其中，限速点信息还可以包括限速点的标识和限速点类型，该限速点的标识可以包括限速点所在设备的标识和/或限速点的编号。该设备的标识可以包括：互联网协议(Internet Protocol，IP)地址和/或网元标识(NE Identity，NEID)。该挂接关系信息基于物理链路上的每个限速点的标识以及每个限速点的限速点类型建立。前述挂接关系信息可以由关系图或者关系表的方式表征。
前述限速点类型包括主路限速点或支路限速点。假设一条物理链路包括m条成员链路，m为正整数，该主路限速点为m条成员链路共同经过的限速点，该支路限速点为m条成员链路分别经过的限速点。以图1为例，对于物理链路1031，该物理链路包括2条成员链路，假设该2条成员链路分别为经过限速点A、B和C1的第一成员链路，以及分别经过限速点A、B和C2的第二成员链路，则主路限速点为限速点A和B，支路限速点分别为第一成员链路经过的限速点C1，以及第二成员链路经过的限速点C2。
图4是本申请实施例提供的一种示意性的挂接关系示意图。该物理链路的挂接关系信息包括：与m条成员链路对应的m个第一分支X(例如m个第一分支X与m条成员链路一一对应)，以及与m个第一分支X连接的一个第二分支Y。每个第一分支X用于记录对 应的成员链路上的限速点类型为支路限速点的限速点的标识x1，不同第一分支X上记录的限速点的标识的个数可以相同也可以不同。第二分支Y用于记录物理链路上的限速点类型为主路限速点的限速点的标识y1。其中，第二分支Y位于第一分支X之后，如此在后续基于挂接关系图进行限速时，是先基于第一分支进行限速，再基于第二分支进行限速，最后限速完成的切片报文从发送端网元输出。
图5是图1所示的物理链路1031的挂接关系信息示意图。如图5所示，物理链路1031包括经过限速点A、B和C1的第一成员链路，以及分别经过限速点A、B和C2的第二成员链路。则对应的挂接关系信息包括：与2条成员链路对应的2个第一分支，以及与2个第一分支连接的一个第二分支Y。假设与第一成员链路和第二成员链路对应的第一分支分别为第一分支X1和X2，则第一分支X1用于记录第一成员链路上的限速点的标识C1，第一分支X2用于记录第二成员链路上的限速点的标识C2，第二分支Y用于记录物理链路上的限速点类型为主路限速点的限速点的标识A和B。
在一种可选实现方式中，在物理链路的挂接关系信息中，对于第一分支和第二分支中的每个分支，该分支记录的限速点的标识可以随机排列。
在另一种可选方式中，对于第一分支和第二分支中的每个分支，该分支上的限速点的标识可以按照预设规则排列。例如，当发送端网元在网元内对切片报文进行限速时，实际限速效果不仅受限速点的带宽的影响，还受限速点的开销比特数的影响，因此可以基于限速点的开销比特数以及带宽进行分支中限速点的标识的排列。该分支上的限速点的标识满足以下至少一种情况：
第一种情况，当两个限速点的带宽相同，且开销比特数不同时，该两个限速点的标识在分支中按照对应的开销比特数由大到小的顺序排列。由于在经过限速点的切片报文的长度不变，且限速点的带宽不变时，限速点的实际限速效果由其添加的开销比特数决定，添加的开销比特数越大，对切片报文的传输速率的限制越大。因此当两个限速点的带宽相同，且开销比特数不同时，将对应的开销比特数较大的限速点的标识排在前面。如此在进行限速时，优先采用前一个限速点的带宽和开销比特数进行限速，使得限速后的切片报文在到达后一个限速点时，已经满足后一个限速点的限速要求，降低切片报文在后一个限速点的通过时延。
第二种情况，当两个限速点的开销比特数相同，且带宽不同时，该两个限速点的标识在分支中按照对应的带宽由小到大的顺序排列。由于在经过限速点的切片报文的长度不变，且限速点的开销比特数不变时，限速点的实际限速效果由其带宽决定，带宽越小，对切片报文的传输速率的限制越大。因此当两个限速点的开销比特数相同，且带宽不同时，将对应的带宽较小的限速点的标识排在前面。如此在进行限速时，优先采用前一个限速点的带宽和开销比特数进行限速，使得限速后的切片报文在到达后一个限速点时，已经满足后一个限速点的限速要求，降低切片报文在后一个限速点的通过时延。
在本申请实施例中，每个第一分支通常记录对应的成员链路上的所有限速点类型为支路限速点的限速点的标识；第二分支通常记录物理链路上的所有限速点类型为主路限速点的限速点的标识。如此，保证物理链路的挂接关系信息覆盖物理链路上的所有限速点的标识，避免出现限速点的遗漏，影响限速结果。
在实际实现时，对于第一分支和第二分支中的每个分支，还可以将该分支所记录的限 速点的标识进行整合，保留需要整合的限速点中的一个限速点的标识。如此，每个第一分支记录的限速点的标识数量小于或等于对应的成员链路上的限速点类型为支路限速点的限速点的标识总数；第二分支记录的限速点的标识数量小于或等于物理链路上的限速点类型为主路限速点的限速点的标识总数。示例地，当发送端网元在网元内对切片报文进行限速时，实际限速效果不仅受限速点的带宽的影响，还受限速点的开销比特数的影响。因此可以基于限速点的开销比特数以及带宽进行分支中限速点的标识的整合。其中，整合的限速点的标识主要包括以下三类：
第一类，开销比特数相同，且带宽相同的两个限速点的标识。对于记录有该类的两个限速点的标识的分支，由于两个限速点的实际限速效果相同，将两个限速点的标识整合为一个限速点的标识，最终得到更新后的分支。基于该分支进行限速可以达到基于更新前的分支进行限速的相同效果。但是更新后的分支记录的限速点的标识数量更少，复杂度更低。示例地，该整合过程包括：对于第一分支和第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽相同，删除该分支中记录的该两个限速点中一个限速点的标识得到更新后的分支。
第二类，开销比特数相同，且带宽不同的两个限速点的标识。对于记录有该类的两个限速点的标识的分支，由于带宽小的限速点相对于带宽大的限速点已经能够实现有效的限速，因此，可以保留带宽小的限速点的标识，最终得到更新后的分支。基于该分支进行限速可以达到基于更新前的分支进行限速的相同效果。但是更新后的分支记录的限速点的标识数量更少，复杂度更低。示例地，该整合过程包括：对于第一分支和第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽不同，删除该分支中记录的该两个限速点中带宽较大的限速点的标识得到更新后的分支。
第三类，开销比特数不同，且带宽相同的两个限速点的标识。对于记录有该类的两个限速点的标识的分支，由于开销比特大的限速点相对于开销比特小的限速点已经能够实现有效的限速，因此，可以保留开销比特小的限速点的标识，最终得到更新后的分支。基于该分支进行限速可以达到基于更新前的分支进行限速的相同效果。但是更新后的分支记录的限速点的标识数量更少，复杂度更低。示例地，该整合过程包括：对于第一分支和第二分支中的每个分支，当两个限速点的开销比特数不同，且带宽相同，删除该分支中记录的该两个限速点中开销比特数较小的限速点的标识得到更新后的分支。
更新后的分支结构简洁，可降低基于该分支进行限速时的复杂度。
需要说明的是，前述限速点信息还包括开销比特数。如此，在发送端网元可以通过获取的限速点信息中的开销比特数以及带宽来进行分支上的限速点的标识的排列和/或整合。
B2、发送端网元基于物理链路上的多个限速点的带宽，按照挂接关系信息中各个限速点的连接关系，对待在物理链路上传输的切片报文进行多级限速。
如前所述，挂接关系信息中记录的每个限速点的标识均对应一个带宽。发送端网元按照挂接关系信息中各个分支上限速点的标识的排列顺序，采用限速点的标识所对应的带宽依次进行待在物理链路上传输的切片报文的限速。其中，一个成员链路对应的分支上的限速可以包括一级限速或多级限速，该多级限速指的是每个切片报文至少经过两级限速，即采用至少两个限速点的带宽对其进行限速。
若物理链路的挂接关系信息如图5所示，则对于待在物理链路上传输的切片报文，若 该切片报文是待经过第一成员链路的切片报文，采用限速点C1的带宽、限速点A的带宽和限速点B的带宽依次对该切片报文进行限速；若该切片报文是待经过第二成员链路的切片报文，采用限速点C2的带宽、限速点A的带宽以及限速点B的带宽依次对该切片报文进行限速。
示例地，对于该挂接关系信息中记录的每个限速点，采用该限速点的带宽进行报文的限速的过程包括：将该限速点的带宽与开销比特数对应的带宽之差作为阈值带宽，控制待经过该限速点的切片报文的单位时间内的通过速度小于该限速点的阈值带宽。
本申请实施例中，可以采用以下两种可选方式来进行切片报文的具体限速。在第一种可选方式中，发送端网元基于获取的物理链路上的多个限速点的限速点信息，建立限速点的标识与带宽的对应关系，将该对应关系存储在预设存储空间中。在进行多级限速时，在遍历到挂接关系信息中的一个限速点的标识后，采用该标识查询限速点的标识与带宽的对应关系，基于查询得到的带宽进行切片报文的限速。由于该对应关系另存在预设存储空间中，便于限速点信息更新后及时更新该对应关系，保证限速的可靠性。在第二种可选方式中，挂接关系信息记录有限速点的标识和每个限速点的标识所对应的带宽。在进行多级限速时，在遍历到挂接关系信息中的一个限速点的标识后，采用该标识对应的带宽进行切片报文的限速。由于限速点对应的标识直接记录在挂接关系信息中，无需在限速过程中进行限速点的标识对应的带宽的查询，减少限速过程的复杂度，降低限速时延。
在本申请实施例中，发送端网元可以通过硬件限速或软件限速的方式，实现对切片报文的多级限速。
第一、硬件限速方式：
对于每条物理链路，发送端网元中预先配置有预设个数的硬件的限速器。该限速器为限速电路或限速芯片，该预设个数通常大于或等于该物理链路的限速点的总数。发送端网元在获取挂接关系信息后，按照挂接关系信息中各个限速点的标识的排列顺序，在预设个数的限速器中选择多个限速器，并建立多个限速器的连接网络。该多个限速器与挂接关系信息中多个限速点的标识一一对应。该建立多个限速点的连接网络的过程称之为挂接。挂接完成后，该多个限速器的连接网络的入口与发送端网元中的切片模块的输出端连接，该连接网络的出口与发送端网元的用于向物理链路输出切片报文的物理端口连接。其中，切片模块用于对业务报文执行报文切片流程。每个限速器用于采用对应的限速点的带宽对经过该限速器的切片报文进行限速。其中，该限速过程指的是限速器控制经过该限速器的切片报文的单位时间内的通过速度小于对应的限速点的带宽，该切片报文的单位时间内的通过速度可以由该切片报文的长度与预设的报文传输速率的乘积表示。如前所述，由于每个限速点允许通过的切片报文还受限速点添加的开销比特数的影响，因此，每个限速器可以模拟对应限速点的实际限速情况来进行限速。则该限速过程包括：限速器将限速点的带宽与开销比特数对应的带宽之差作为阈值带宽，控制经过该限速器的切片报文的单位时间内的通过速度小于对应的限速点的阈值带宽。其中，开销比特数对应的带宽为该开销比特数与预设的报文传输速率的乘积。
图6是图5所示的物理链路的挂接关系信息所对应的限速器的连接网络示意图。如图6所示，发送端网元在获取挂接关系信息后，按照挂接关系信息中各个限速点的标识的排列顺序，建立4个限速器的连接网络。该4个限速器分别为限速器1-4。该4个限速器分别与 挂接关系信息中4个限速点的标识C1、C2、A和B一一对应。其中，与限速点C1、C2对应的限速器1和2分别连接有与限速点A对应的限速器3，该限速器3还连接有与限速点B对应的限速器4。其中，限速器1和2为连接网络的入口，该限速器4为连接网络的出口。
第二、软件限速方式：
对于每条物理链路，发送端网元在获取挂接关系信息后，按照挂接关系信息中各个限速点的标识的排列顺序，分配多个软件的限速器(即分配多个软件的用于限速的资源)，并建立该多个限速器的连接网络。该多个限速器与挂接关系信息中多个限速点的标识一一对应。该建立多个限速点的连接网络的过程称之为挂接。挂接完成后的软件限速器的连接网络与前述硬件的限速器的连接网络的功能一致，各个限速器的连接方式与功能对应参考前述硬件限速方式中的限速器的连接方式和功能。
前述硬件限速方式相对于软件限速方式的限速可靠性更高，限速器易于管理和维护。
需要说明的是，对于一条物理链路，当对应的挂接关系信息中包括多个第一分支时，需要设置一个调度器，该调度器可以为硬件调度器或软件调度器，该调度器对应该多个第一分支与一个第二分支的连接点，该调度器用于对来自不同第一分支对应限速器的切片报文进行调度，以防止切片报文产生拥塞。示例地，前述调度器为轮询(Round Robin，RR)调度器。
S303、发送端网元更新物理链路上的多个限速点中至少一个限速点的限速点信息。
多个限速点的限速点信息可能会由于环境变化或其他因素的变化而产生变化。例如，限速点为空口，由于该限速点所处环境变化，如天气的阴晴变化，导致该限速点的带宽变化。又例如，某一限速点所在硬件设备故障后，进行了硬件设备的更换。从而导致该某一限速点更新，相应的限速点信息，如带宽和限速点的标识也产生变化。发送端网元需要获取变化后的限速点信息，以保证限速效果。
本申请实施例中，发送端网元获取更新的限速点信息的方式有多种，本申请实施例以以下两种为例进行说明：
第一种方式，发送端网元在多个限速点中的任一限速点信息变化时，获取该任一限速点的限速点信息。
发送端网元获取该任一限速点的限速点信息的过程可以参考前述S301的发送端网元获取物理链路上的一个限速点的限速点信息的过程。例如，接收监控设备上报的该任一限速点的限速点信息，或者，通过人工配置的方式获取该任一限速点的限速点信息。
发送端网元在多个限速点中的任一限速点信息变化时，获取该任一限速点的限速点信息。如此可以保证发送端网元对变化的限速点信息的实时获取，保证限速的及时准确。
第二种方式，发送端网元周期性获取多个限速点的限速点信息。
在每个获取周期，发送端网元获取多个限速点的限速点信息的过程可以参考前述S301的发送端网元获取物理链路上的多个限速点的限速点信息的过程。例如，接收监控设备上报的该多个限速点的限速点信息，或者，通过人工配置的方式获取该多个限速点的限速点信息。
S304、发送端网元基于更新后的多个限速点的带宽，对待在物理链路上传输的切片报文进行限速。
发送端网元基于更新后的多个限速点的带宽，对待在物理链路上传输的切片报文进行 限速的过程可以参考前述S302中的限速过程。
其中，参考前述B2，若发送端网元建立有限速点的标识与带宽的对应关系，发送端网元先基于更新后的多个限速点的限速点信息，更新限速点的标识与带宽的对应关系；再基于该更新后的对应关系，查询得到更新后的多个限速点的带宽以对待在物理链路上传输的切片报文进行限速。
若挂接关系信息记录有限速点的标识以及每个限速点的标识所对应的带宽，发送端网元先基于更新后的多个限速点的限速点信息，更新挂接关系信息；然后基于更新后的挂接关系信息对待在物理链路上传输的切片报文进行限速。
相关技术中，对于一个成员链路，发送端网元需要进行切片报文的一次限速，用于限速的带宽是固定带宽。也即是，在工作人员设置该带宽之后，该带宽不再变更。如此导致用于限速的带宽的灵活性较低。
而本申请实施例中，发送端网元可以更新物理链路上的多个限速点中至少一个限速点的限速点信息，例如周期性更新或实时更新限速点的信息。如此，用于限速的多个带宽是动态可变的。该多个带宽根据实际情况调整，从而保证用于限速的多个带宽能够有效适配不同的限速环境，提高用于限速的带宽的灵活性。
S305、发送端网元将限速后的切片报文发送到物理链路上。
在本申请实施例中，发送端网元在执行多级限速后，如前述S302或S304之后，通过发送端网元上的物理接口输出至物理链路。此时经过物理链路的切片报文在尽量保证切片报文的最大净荷带宽的基础上进行传输，从而提高带宽利用率。
综上所述，本申请实施例中，发送端网元通过获取的物理链路上的多个限速点的带宽，对待在该物理链路上传输的切片报文进行限速。如此，切片报文在经过物理链路之前，已经在发送端网元处按照各个限速点的带宽进行了限速。因此限速后的切片报文在实际在物理链路上传输时已经满足了各个限速点的带宽要求。在此基础上，发送端网元直接获取限速点本身的带宽，并基于获取的带宽进行限速。无需人工确定每个限速点的限速带宽，减少人工成本，简化了用于限速的带宽的获取过程，提高了用于限速的带宽的获取效率。
本申请实施例中，通信系统还支持带宽预测功能，通过提前预测带宽的变化可以实现灵活限速。该带宽预测功能包括以下步骤：
C1、发送端网元接收预测设备发送的带宽预测信息。
该预测设备与多条物理链路中的至少一条物理链路对应，该预测设备用于预测对应的物理链路上监控到的限速点的变化后的带宽。该预测设备的设置方式可以与前述S301中监控设备的设置方式相同或相近。例如，聚合链路中每条物理链路包括至少一个硬件设备，该至少一个硬件设备包括前述预测设备。在第一种情况下，该至少一个硬件设备中的限速点所在硬件设备为预测设备，任一预测设备在对应的物理链路上监控到的限速点包括位于该任一预测设备上的限速点和/或与该任一预测设备连接的限速点。在第二种情况下，该至少一个硬件设备中的指定硬件设备为预测设备，该预测设备在对应的物理链路上监控到的限速点包括该对应的物理链路上的一个或多个限速点。其中，预测设备监控到的限速点为对应的物理链路上的带宽可变的限速点，例如空口。
其中，预测设备和前述监控设备可以是相同的设备，也可以是不同的设备。预测设备中预先配置有所监控的限速点的带宽变化条件。预测设备在监控到某一限速点满足带宽变 化条件后，预测该某一限速点的变化后的带宽，并向发送端网元发送包括预测带宽的带宽预测信息。该预测带宽为该预测设备在确定对应的物理链路上监控到的某一限速点满足带宽变化条件后，预测的该某一限速点的变化后的带宽。
示例地，该带宽变化条件为：限速点所处环境开始产生预设变化，或者，限速点所处环境在预设时长后产生预设变化；或者限速点的带宽在预设时长后重置等。其中，当限速点为空口时，该预设变化包括：变化幅度大于预设温差的温度变化，和/或，变化幅度大于预设湿度差的湿度变化等。
在一种可选方式中，预测设备用于通过自主监测的方式确定限速点是否满足带宽变化条件；在另一种可选方式中，预测设备用于接收其他设备发送的触发消息，该触发消息用于指示该预测设备监控到的限速点是否满足带宽变化条件。例如，预测设备包括环境监测模块，预测设备通过该环境监测模块监控限速点的环境，以确定限速点是否满足带宽变化条件；又例如，预测设备设置有通信模块，该通信模块用于接收设置在限速点附近的环境监测设备发送的触发消息。
本申请实施例中，预测设备在监控到某一限速点满足带宽变化条件后，可以通过多种方式预测该某一限速点的变化后的带宽。例如，若该带宽变化条件为：限速点所处环境开始产生预设变化，或者，限速点所处环境在预设时长后产生预设变化；预测设备根据环境(如温度和/或湿度)与带宽的变化关系，确定变化后的带宽。若该带宽变化条件为：限速点的带宽在预设时长后重置，预测设备预先获取重置后的带宽作为预测带宽。
需要说明的是，预测设备可以重复发送该带宽预测信息，以保证发送端网元能够收到该带宽预测信息，提高带宽预测信息上报的可靠性。
C2、发送端网元基于预测带宽，对物理链路上传输的切片报文中待经过某一限速点的切片报文进行限速。
发送端网元基于预测带宽，对物理链路上传输的切片报文中待经过某一限速点的切片报文进行限速的过程可以参考前述S302中的限速过程。
其中，参考前述B2，若发送端网元建立有限速点的标识与带宽的对应关系。发送端网元先基于该限速带宽，更新限速点的标识与带宽的对应关系；再基于该更新后的对应关系，查询得到多个限速点的带宽以对待在物理链路上传输的切片报文进行限速。其中，查询得到的多个限速点的带宽包括该预测带宽，待在物理链路上传输的切片报文包括待经过该某一限速点的切片报文。
若挂接关系信息记录有限速点的标识以及每个限速点的标识所对应的带宽，发送端网元先基于该限速带宽，更新挂接关系信息；然后基于更新后的挂接关系信息对待在物理链路上传输的切片报文进行限速。其中，挂接关系信息包括该预测带宽，待在物理链路上传输的切片报文包括待经过该某一限速点的切片报文。
本申请实施例中，当发送端网元支持带宽预测功能时，通预测设备提前预测即将生效的带宽，并通知发送端网元。由该发送端网元进行响应，提前进行该网元内的带宽调整，从而实现无损的带宽切换。尤其对于一些优先级较高的业务报文，有效防止因为限速失效所导致的切片报文丢失，从而避免随机丢包现象。
相关技术中，由于采用经过成员链路的切片报文的最小长度转化得到的带宽来确定上限带宽，导致较长的切片报文经过成员链路的实际净荷带宽(即切片报文在单位时间实际 通过一个限速点的净荷)远远小于其最大净荷带宽。如此长度较长的切片报文(即长度大于该最小长度的切片报文)在单位时长内通过的净荷占该切片报文的长度的比例(该比例称为通过率)较低，使得长度较长的切片报文在该成员链路的通信损耗较大(例如，达到30％以上)。
本申请实施例中，发射端网元直接采用各个限速点的带宽以及开销比特数进行不同长度的切片报文的限速。因此，对于不同长度的切片报文，均能在尽量保证其在最大净荷带宽或者接近最大净荷带宽的情况下通过，从而减少通信损耗。
但是，如果某一限速点对应的切片报文的长度和限速点的开销比特数之和所对应的带宽大于该限速点的带宽。在发送端网元处，容易在采用该限速点的带宽以及开销比特数进行限速时，出现拥塞现象。严重的拥塞会产生切片报文的丢失，从而产生随机丢包的可能性。
在本申请实施例中，发送端网元通过进行流量监控，减少拥塞现象，从而避免随机丢包。示例地，本申请实施例的发送端网元还支持流控反压的方式。流控反压指的是通过反馈的方式实现切片报文的流量控制。本申请实施例中，该流控反压的过程包括：
对于挂接关系信息记录的每个限速点，当待经过该限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽大于限速点的带宽，发送端网元通过丢弃业务报文的方式降低待经过限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽。该业务报文用于切片得到切片报文。其中，该限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽是基于预设的转换算法计算得到的。示例地，该转换算法满足：w＝v(h1+h2)；其中，w为限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽，v为预设的报文传输速率，h1为切片报文的长度，h2为限速点的开销的长度(如开销比特数或开销字节数)。该转换算法中，h1和h2的单位相同，或者经过换算后单位相同。
参考前述S302，在采用硬件限速方式或软件限速方式进行限速时，多个限速器的连接网络的入口与切片模块连接。假设第一限速器是该多个限速器中的任一限速器，第一限速点是该第一限速器对应的限速点。则第一限速器还用于检测切片报文的长度和第一限速点的开销比特数之和所对应的带宽是否大于第一限速点的带宽。当检测得到切片报文的长度和第一限速点的开销比特数之和所对应的带宽大于第一限速点的带宽，在第一种可选方式中，第一限速器向第一限速器的上游反馈反压信息(若第一限速器是多个限速器的连接网络的入口，则该第一限速器向切片模块反馈反压信息；若第一限速器不是多个限速器的连接网络的入口，则该第一限速器向上一级限速器反馈反压信息)，该反压信息用于指示切片报文的长度和第一限速点的开销比特数之和所对应的带宽大于第一限速点的带宽。该反压信息按照挂接关系信息中限速点的标识的排列顺序相反的顺序由各个限速器向前反馈，直至到达切片模块；切片模块通过丢弃待分片的业务报文，从而减少单位时长内发出的切片报文的数量，即降低报文流量，从而降低到达第一限速器的切片报文的长度和限速点的开销比特数之和所对应的带宽。
如图6所示，假设第一限速器为限速器4，其向该限速器4的上一级限速器：限速器3发送反压信息，由限速器3向该限速器3的上一级限速器：限速器1和/或限速器2发送反压信息，限速器1和/或限速器2向切片模块发送反压信息。
在第二种可选方式中，多个限速器还分别和切片模块连接。当检测得到切片报文的长 度和第一限速点的开销比特数之和所对应的带宽大于第一限速点的带宽，第一限速器向切片模块反馈反压信息。相对于第一种可选方式，该种方式下，第一限速器到切片模块的时延较小，能够实现报文流量的快速调整。但是发送端网元内部的连接关系较为复杂，管理成本较高。
在本申请实施例中，发送端网元可以根据第一限速器的拥塞情况，来确定丢弃的业务报文的数量。拥塞情况越严重，丢弃的业务报文的数量越多。示例地，反压信息包括拥塞等级，该拥塞等级用于反映拥塞的情况的严重性。第一限速器可以将切片报文的长度和第一限速点的开销比特数之和所对应的带宽与第一限速点的带宽做差，采用得到的目标差值查询预设的拥塞等级与差值范围对应关系。将该目标差值所在差值范围对应的拥塞等级携带在反压信息中。发送端网元基于接收的反压信息中的拥塞等级，查询拥塞等级与丢弃数量的对应关系，得到的目标丢弃数量，并丢弃该目标丢弃数量的业务报文。示例地，业务报文携带有优先级，发送端网元按照优先级由低到高的顺序进行业务报文的丢弃，如此在保证降低拥塞的前提下，尽量减少优先级高的业务报文的丢弃。
由于发送端网元通过直接丢弃业务报文的方式来进行流量控制，避免切片报文的丢失所导致的随机丢包，尤其减少了高优先级的业务报文的丢包，从而保证了通信过程的可靠性。并且在发送端网元采用流控反压的方式，使得各种长度的切片报文，都能尽力传输，有效提高了切片报文的传输效率。
本申请前述实施例中，通信系统中的发送端网元和接收端网元是相对而言的。在具体应用中，网元可能既是发送端网元也是接收端网元。
需要说明的是，本申请实施例提供的通信方法步骤的先后顺序可以进行适当调整，步骤也可以根据情况进行相应增减，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化的方法，都应涵盖在本申请的保护范围之内，因此不再赘述。此外，装置实施例的有益效果和方法对应步骤的有益效果类似，下面不再赘述。
图7是本申请实施例提供的一种通信装置40的结构示意图。该装置40应用于发送端网元，该装置40包括第一获取模块401和第一限速模块402。其中，第一获取模块401，用于对于通过PLA捆绑的多条物理链路中每条物理链路，获取物理链路上的多个限速点的限速点信息。该限速点信息包括带宽。第一限速模块402，用于基于物理链路上的多个限速点的带宽，对待在物理链路上传输的切片报文进行限速。第一获取模块401，用于接收监控设备发送的监控信息，监控设备与多条物理链路中的至少一条物理链路对应，监控信息包括监控设备在对应的物理链路上监控到的限速点的限速点信息。
可选地，多条物理链路中每条物理链路包括至少一个硬件设备，至少一个硬件设备中限速点所在硬件设备为监控设备，任一监控设备在对应的物理链路上监控到的限速点包括位于任一监控设备上的限速点和/或与任一监控设备连接的限速点；第一获取模块401，用于：接收每条物理链路上的监控设备发送的监控信息。
在一种可选实现方式中，第一获取模块401，用于：接收带宽设置指令，带宽设置指令包括多条物理链路中一个或多个限速点的限速点信息。
可选地，如图7所示，该装置还包括：第二获取模块403。该第二获取模块403用于：在多个限速点中的任一限速点信息变化时，获取任一限速点的限速点信息；和/或，周期性获取多个限速点的限速点信息。
可选地，如图7所示，该装置还包括：接收模块404和第二限速模块405。其中，接收模块404，用于接收预测设备发送的带宽预测信息。该预测设备与多条物理链路中的至少一条物理链路对应。该带宽预测信息包括预测带宽。该预测带宽为预测设备预测的预测设备对应的物理链路上的任一限速点的带宽变化后的带宽。第二限速模块405，用于基于预测带宽，对待在物理链路上传输的切片报文中待经过任一限速点的切片报文进行限速。
在一种可选实现方式中，通信装置40中的第一限速模块402可以基于挂接关系信息来进行限速。示例地，第一限速模块402，用于：获取挂接关系信息，该挂接关系信息用于描述物理链路上各个限速点的连接关系。该挂接关系信息基于物理链路上的每个限速点的标识以及每个限速点的限速点类型建立。基于物理链路上的多个限速点的带宽，按照挂接关系信息中各个限速点的连接关系，对待在物理链路上传输的切片报文进行多级限速。
可选地，如图7所示，装置40还包括：处理模块406，用于对于挂接关系信息记录的每个限速点。当待经过限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽大于限速点的带宽，通过丢弃业务报文的方式降低待经过限速点的切片报文的长度和限速点的开销比特数之和所对应的带宽。该业务报文用于切片得到切片报文。
示例地，物理链路包括m条成员链路，m为正整数。该限速点信息还包括：限速点的标识和限速点类型。该限速点类型包括主路限速点或支路限速点。其中，主路限速点为m条成员链路共同经过的限速点，支路限速点为m条成员链路分别经过的限速点。
在一种可选示例中，物理链路的挂接关系信息包括的信息参见图4所示的描述，即第一分支和第二分支，在此不再赘述。示例地，对于第一分支和第二分支中的每个分支，分支上的限速点的标识满足以下至少一种：当两个限速点的带宽相同，且开销比特数不同时，两个限速点的标识在分支中按照对应的开销比特数由大到小的顺序排列；当两个限速点的开销比特数相同，且带宽不同时，两个限速点的标识在分支中按照对应的带宽由小到大的顺序排列。
可选地，对于第一分支和第二分支中的每个分支，处理模块406用于当两个限速点的开销比特数相同，且带宽不同，删除分支中记录的两个限速点中带宽较大的限速点的标识得到更新后的分支；或者，当两个限速点的开销比特数不同，且带宽相同，删除该分支中记录的该两个限速点中开销比特数较小的限速点的标识得到更新后的分支；或者，当两个限速点的开销比特数相同，且带宽相同，删除该分支中记录的该两个限速点中一个限速点的标识得到更新后的分支。
图8是本申请又一实施例提供的一种通信装置50的结构示意图。如图8所示，该通信装置50包括：处理芯片501，该处理芯片501包括处理电路5011及通信接口5012。该处理电路用于本申请前述实施例中发送端所执行的通信方法。该处理电路5012可以为处理芯片或现场可编程逻辑门阵列(Field Programmable Gate Array，FPGA)。该处理芯片可以为集成电路(Application Specific Integrated Circuit，ASIC)芯片；该通信接口5012用于供处理电路5011与其他装置进行通信。例如，该通信接口5012用于供处理电路5011通过聚合链路与接收端网元进行通信。该通信接口5012包括输入接口和输出接口。通信接口5012可以是以下器件的任一种或任一种组合：网络接口(例如以太网接口)或无线网卡等。
在一种可选实现方式中，该处理电路5012包括缓存结构，如FPGA或ASIC芯片内部的存储结构，用于缓存限速点信息。在另一种可选实现方式中，该通信装置501还包括： 存储器，用于缓存限速点信息。例如，该存储器为快闪存储器(flash memory)。
本申请实施例提供一种通信系统，该通信系统包括发送端网元、接收端网元和以及通过PLA捆绑的多条物理链路，该发送端网元包括本申请实施例提供的通信装置，如通信装置40或通信装置50。该通信系统的结构可参考图1或图2。
在本申请中，术语“第一”、“第二”和“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。术语“至少一个”表示1个或多个，术语“多个”指两个或两个以上，除非另有明确的限定。A参考B，指的是A与B相同或者A为B的简单变形。
需要说明的是：上述实施例提供的通信装置在执行通信方法时，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。另外，上述实施例提供的通信装置与通信方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本申请的可选实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (26)

  1. 一种通信方法，其特征在于，应用于发送端网元，所述方法包括：
    对于通过物理链路聚合PLA捆绑的多条物理链路中每条物理链路，获取所述物理链路上的多个限速点的限速点信息，所述限速点信息包括带宽；
    基于所述物理链路上的多个限速点的带宽，对待在所述物理链路上传输的切片报文进行限速。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：在所述多个限速点中的任一限速点信息变化时，获取所述任一限速点的限速点信息；和/或，周期性获取所述多个限速点的限速点信息。
  3. 根据权利要求1或2所述的方法，其特征在于，所述获取所述物理链路上的多个限速点的限速点信息，包括：接收监控设备发送的监控信息，所述监控设备与所述多条物理链路中的至少一条物理链路对应，所述监控信息包括所述监控设备在对应的物理链路上监控到的限速点的限速点信息。
  4. 根据权利要求3所述的方法，其特征在于，所述多条物理链路中每条物理链路包括至少一个硬件设备，所述至少一个硬件设备中限速点所在硬件设备为所述监控设备，任一监控设备在对应的物理链路上监控到的限速点包括位于所述任一监控设备上的限速点和/或与所述任一监控设备连接的限速点；
    所述接收监控设备发送的监控信息，包括：接收所述每条物理链路上的监控设备发送的监控信息。
  5. 根据权利要求1至4任一所述的方法，其特征在于，所述获取所述物理链路上的多个限速点的限速点信息，包括：接收带宽设置指令，所述带宽设置指令包括所述多条物理链路中一个或多个限速点的限速点信息。
  6. 根据权利要求1至5任一所述的方法，其特征在于，所述方法还包括：
    接收预测设备发送的带宽预测信息，所述预测设备与所述多条物理链路中的至少一条物理链路对应，所述带宽预测信息包括预测带宽，所述预测带宽为所述预测设备在对应的物理链路上监控到的某一限速点满足带宽变化条件后，预测的所述某一限速点的变化后的带宽；
    基于所述预测带宽，对待在所述物理链路上传输的切片报文中待经过所述某一限速点的切片报文进行限速。
  7. 根据权利要求1至6任一所述的方法，其特征在于，所述基于所述物理链路上的多个限速点的限速点信息，对待在所述物理链路上传输的切片报文进行限速，包括：
    获取挂接关系信息，所述挂接关系信息用于描述所述物理链路上各个限速点的连接关系，所述挂接关系信息基于所述物理链路上的每个限速点的标识以及所述每个限速点的限速点类型建立；
    基于所述物理链路上的多个限速点的带宽，按照所述挂接关系信息中各个限速点的连接关系，对待在所述物理链路上传输的切片报文进行多级限速。
  8. 根据权利要求7所述的方法，其特征在于，所述方法还包括：对于所述挂接关系信息记录的每个限速点，当待经过所述限速点的切片报文的长度和所述限速点的开销比特数之和所对应的带宽大于所述限速点的带宽，通过丢弃业务报文的方式降低待经过所述限速点的切 片报文的长度和所述限速点的开销比特数之和所对应的带宽，所述业务报文用于切片得到切片报文。
  9. 根据权利要求7或8所述的方法，其特征在于，所述物理链路包括m条成员链路，m为正整数，所述限速点信息还包括：限速点的标识和限速点类型，所述限速点类型包括主路限速点或支路限速点，所述主路限速点为所述m条成员链路共同经过的限速点，所述支路限速点为所述m条成员链路分别经过的限速点。
  10. 根据权利要求9所述的方法，其特征在于，所述物理链路的挂接关系信息包括：与所述m条成员链路对应的m个第一分支，以及与所述m个第一分支连接的一个第二分支；
    每个第一分支用于记录对应的成员链路上的限速点类型为支路限速点的限速点的标识，所述第二分支用于记录所述物理链路上的限速点类型为主路限速点的限速点的标识。
  11. 根据权利要求10所述的方法，其特征在于，对于所述第一分支和所述第二分支中的每个分支，所述分支上的限速点的标识满足以下至少一种：
    当两个限速点的带宽相同，且开销比特数不同时，所述两个限速点的标识在所述分支中按照对应的开销比特数由大到小的顺序排列；
    当两个限速点的开销比特数相同，且带宽不同时，所述两个限速点的标识在所述分支中按照对应的带宽由小到大的顺序排列。
  12. 根据权利要求10所述的方法，其特征在于，所述方法还包括：
    对于所述第一分支和所述第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽不同，删除所述分支中记录的所述两个限速点中带宽较大的限速点的标识得到更新后的分支；
    或者，当两个限速点的开销比特数不同，且带宽相同，删除所述分支中记录的所述两个限速点中开销比特数较小的限速点的标识得到更新后的分支；
    或者，当两个限速点的开销比特数相同，且带宽相同，删除所述分支中记录的所述两个限速点中一个限速点的标识得到更新后的分支。
  13. 一种通信装置，其特征在于，应用于发送端网元，所述装置包括：
    第一获取模块，用于对于通过物理链路聚合PLA捆绑的多条物理链路中每条物理链路，获取所述物理链路上的多个限速点的限速点信息，所述限速点信息包括带宽；
    第一限速模块，用于基于所述物理链路上的多个限速点的带宽，对待在所述物理链路上传输的切片报文进行限速。
  14. 根据权利要求13所述的装置，其特征在于，所述装置还包括：第二获取模块，用于在所述多个限速点中的任一限速点信息变化时，获取所述任一限速点的限速点信息；和/或，用于周期性获取所述多个限速点的限速点信息。
  15. 根据权利要求13或14所述的装置，其特征在于，所述第一获取模块还用于接收监控设备发送的监控信息，所述监控设备与所述多条物理链路中的至少一条物理链路对应，所述监控信息包括所述监控设备在对应的物理链路上监控到的限速点的限速点信息。
  16. 根据权利要求15所述的装置，其特征在于，所述多条物理链路中每条物理链路包括至少一个硬件设备，所述至少一个硬件设备中限速点所在硬件设备为所述监控设备，任一监控设备在对应的物理链路上监控到的限速点包括位于所述任一监控设备上的限速点和/或与所述任一监控设备连接的限速点；
    所述第一获取模块还用于：接收所述每条物理链路上的监控设备发送的监控信息。
  17. 根据权利要求13至16任一所述的装置，其特征在于，所述第一获取模块还用于：接收带宽设置指令，所述带宽设置指令包括所述多条物理链路中一个或多个限速点的限速点信息。
  18. 根据权利要求13至17任一所述的装置，其特征在于，所述装置还包括：
    接收模块，用于接收预测设备发送的带宽预测信息，所述预测设备与所述多条物理链路中的至少一条物理链路对应，所述带宽预测信息包括预测带宽，所述预测带宽为所述预测设备预测的所述预测设备对应的物理链路上的任一限速点的带宽变化后的带宽；
    第二限速模块，用于基于所述预测带宽，对待在所述物理链路上传输的切片报文中待经过所述任一限速点的切片报文进行限速。
  19. 根据权利要求13至18任一所述的装置，其特征在于，所述第一限速模块，用于：
    获取挂接关系信息，所述挂接关系信息用于描述所述物理链路上各个限速点的连接关系，所述挂接关系信息基于所述物理链路上的每个限速点的标识以及所述每个限速点的限速点类型建立；
    基于所述物理链路上的多个限速点的带宽，按照所述挂接关系信息中各个限速点的连接关系，对待在所述物理链路上传输的切片报文进行多级限速。
  20. 根据权利要求19所述的装置，其特征在于，所述装置还包括：处理模块，用于对于所述挂接关系信息记录的每个限速点，当待经过所述限速点的切片报文的长度和所述限速点的开销比特数之和所对应的带宽大于所述限速点的带宽，通过丢弃业务报文的方式降低待经过所述限速点的切片报文的长度和所述限速点的开销比特数之和所对应的带宽，所述业务报文用于切片得到切片报文。
  21. 根据权利要求19或20所述的装置，其特征在于，所述物理链路包括m条成员链路，m为正整数，所述限速点信息还包括：限速点的标识和限速点类型，所述限速点类型包括主路限速点或支路限速点，所述主路限速点为所述m条成员链路共同经过的限速点，所述支路限速点为所述m条成员链路分别经过的限速点。
  22. 根据权利要求21所述的装置，其特征在于，所述物理链路的挂接关系信息包括：与所述m条成员链路对应的m个第一分支，以及与所述m个第一分支连接的一个第二分支；
    每个第一分支用于记录对应的成员链路上的限速点类型为支路限速点的限速点的标识，所述第二分支用于记录所述物理链路上的限速点类型为主路限速点的限速点的标识。
  23. 根据权利要求22所述的装置，其特征在于，对于所述第一分支和所述第二分支中的每个分支，所述分支上的限速点的标识满足以下至少一种：
    当两个限速点的带宽相同，且开销比特数不同时，所述两个限速点的标识在所述分支中按照对应的开销比特数由大到小的顺序排列；
    当两个限速点的开销比特数相同，且带宽不同时，所述两个限速点的标识在所述分支中按照对应的带宽由小到大的顺序排列。
  24. 根据权利要求22所述的装置，其特征在于，所述装置还包括：处理模块，用于：
    对于所述第一分支和所述第二分支中的每个分支，当两个限速点的开销比特数相同，且带宽不同，删除所述分支中记录的所述两个限速点中带宽较大的限速点的标识得到更新后的分支；
    或者，当两个限速点的开销比特数不同，且带宽相同，删除所述分支中记录的所述两个限速点中开销比特数较小的限速点的标识得到更新后的分支；
    或者，当两个限速点的开销比特数相同，且带宽相同，删除所述分支中记录的所述两个限速点中一个限速点的标识得到更新后的分支。
  25. 一种通信装置，其特征在于，所述通信装置包括：
    处理芯片，所述处理芯片包括：处理电路及通信接口，所述处理电路用于执行权利要求1至12任一所述的通信方法；
    所述通信接口用于供所述处理电路与其他装置进行通信。
  26. 一种通信系统，其特征在于，所述通信系统包括发送端网元、接收端网元和以及通过物理链路聚合PLA捆绑的多条物理链路，所述发送端网元包括权利要求25所述的通信装置。
PCT/CN2021/130687 2020-11-27 2021-11-15 通信方法、装置及系统 WO2022111325A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011359748.2A CN114629841B (zh) 2020-11-27 2020-11-27 通信方法、装置及系统
CN202011359748.2 2020-11-27

Publications (1)

Publication Number Publication Date
WO2022111325A1 true WO2022111325A1 (zh) 2022-06-02

Family

ID=81754007

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/130687 WO2022111325A1 (zh) 2020-11-27 2021-11-15 通信方法、装置及系统

Country Status (2)

Country Link
CN (1) CN114629841B (zh)
WO (1) WO2022111325A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101242358A (zh) * 2008-02-20 2008-08-13 华为技术有限公司 一种流量整形方法和装置
US20120044918A1 (en) * 2010-08-19 2012-02-23 Lantiq Deutschland Gmbh Method and Apparatus for Supporting Class of Service Over Bonded Digital Subscriber Line (DSL) Links
CN102377667A (zh) * 2011-10-14 2012-03-14 中兴通讯股份有限公司 板卡及跨板捆绑链路的限速方法
CN102594671A (zh) * 2012-02-08 2012-07-18 中兴通讯股份有限公司 一种对用户进行限速的方法和装置
CN107920026A (zh) * 2016-10-08 2018-04-17 中兴通讯股份有限公司 基于物理端口实现的端口汇聚Trunk分流方法及装置

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0948168A1 (en) * 1998-03-31 1999-10-06 TELEFONAKTIEBOLAGET L M ERICSSON (publ) Method and device for data flow control
CA2308533A1 (en) * 1999-05-14 2000-11-14 Jeffery John Brown Method and apparatus for an optimized digital subscriber loop access multiplexer
JP4283792B2 (ja) * 2005-08-29 2009-06-24 富士通株式会社 帯域制御方法及び伝送装置
US8059540B2 (en) * 2007-05-25 2011-11-15 Futurewei Technologies, Inc. Policy based and link utilization triggered congestion control
CN101478527A (zh) * 2009-01-20 2009-07-08 华为技术有限公司 带宽分配方法和路由设备
CN101567855B (zh) * 2009-06-11 2011-12-21 杭州华三通信技术有限公司 分布式包交换系统和分布式包交换方法
JP2011049891A (ja) * 2009-08-27 2011-03-10 Nec Corp 帯域幅割当装置、ネットワークシステム、帯域幅割当方法
CN102223305A (zh) * 2011-06-17 2011-10-19 华为数字技术有限公司 发送数据的方法和装置、接收数据的方法和装置以及系统
US9882817B2 (en) * 2013-06-26 2018-01-30 Verizon Patent And Licensing Inc. Inter-device policing on network interface devices in LAG configuration
CN103763218B (zh) * 2014-02-18 2020-01-21 网神信息技术（北京）股份有限公司 互联网带宽控制方法、装置和存储介质
CN106453111B (zh) * 2015-08-11 2020-12-22 中兴通讯股份有限公司 基于聚合链路的流量管理方法及装置
CN105898799B (zh) * 2015-10-30 2019-04-30 法法汽车(中国)有限公司 基于信号强度的移动网络的多链路带宽分配方法及装置
CN105763466B (zh) * 2016-03-10 2019-08-06 新华三技术有限公司 一种流量控制方法和装置
US10122651B2 (en) * 2016-08-31 2018-11-06 Inspeed Networks, Inc. Dynamic bandwidth control
CN106357547A (zh) * 2016-09-08 2017-01-25 重庆邮电大学 一种基于流切片的软件定义网络拥塞控制算法
CN106685846B (zh) * 2016-12-29 2020-01-17 北京华为数字技术有限公司 流量控制方法和装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101242358A (zh) * 2008-02-20 2008-08-13 华为技术有限公司 一种流量整形方法和装置
US20120044918A1 (en) * 2010-08-19 2012-02-23 Lantiq Deutschland Gmbh Method and Apparatus for Supporting Class of Service Over Bonded Digital Subscriber Line (DSL) Links
CN102377667A (zh) * 2011-10-14 2012-03-14 中兴通讯股份有限公司 板卡及跨板捆绑链路的限速方法
CN102594671A (zh) * 2012-02-08 2012-07-18 中兴通讯股份有限公司 一种对用户进行限速的方法和装置
CN107920026A (zh) * 2016-10-08 2018-04-17 中兴通讯股份有限公司 基于物理端口实现的端口汇聚Trunk分流方法及装置

Also Published As

Publication number Publication date
CN114629841A (zh) 2022-06-14
CN114629841B (zh) 2023-05-16

Similar Documents

Publication Publication Date Title
US8908522B2 (en) Transmission rate control
KR101748750B1 (ko) Ｓｄｎ 기반 무선 가상 네트워크 환경에서 트래픽 변화를 고려한 대역폭 제어 장치 및 방법
US8964739B1 (en) Self-healing data transmission system and method to achieve deterministic and lower latency
WO2020258746A1 (zh) 数据传输处理方法、装置和通信设备
CN111131450A (zh) 文件上传方法、重组方法及装置
US11671368B2 (en) Congestion control method, apparatus, and system, and computer storage medium
KR20130099201A (ko) 다채널 패킷 송신을 위한 시스템 및 방법
CN117880092A (zh) 网络参数配置方法、装置、计算机设备以及存储介质
CN111181873A (zh) 数据发送方法、装置、存储介质和电子设备
CN107343231B (zh) 非对称缓存与转发模块及由其组成的分组交换节点及网络
EP3682665B1 (en) Packet latency reduction in mobile radio access networks
US8737416B2 (en) Access link aggregator
WO2010072122A1 (zh) 一种传输报文的方法、端口和系统
CN112005528B (zh) 一种数据交换方法、数据交换节点及数据中心网络
WO2022111325A1 (zh) 通信方法、装置及系统
CN115914887A (zh) 一种业务数据的传输方法及相关设备
CN112003685A (zh) 一种灵活以太网报文转发方法及装置
CN115314442B (zh) 拥塞控制和基于Group的限速限窗装置及方法、限速限窗方法
CN114598745B (zh) 通信方法及装置
KR102370114B1 (ko) 지능형 네트워크 관리 시스템의 정보묶음 생성관리 장치 및 그 방법
EP3800831B1 (en) Link group configuration method and apparatus
CN114339889A (zh) 一种无线链路动态选择方法及系统
WO2020063300A1 (zh) 一种数据传输方法、通信设备及存储介质
Kim et al. Toward Enabling Performance-Guaranteed Networking in Next-Generation Cellular Networks
CN113810305B (zh) 报文转发方法、装置、转发节点以及计算机可读存储介质

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21896826

Country of ref document: EP

Kind code of ref document: A1