WO2018040673A1 - 一种用于处理低延迟业务流的方法和装置 - Google Patents

一种用于处理低延迟业务流的方法和装置 Download PDF

Info

Publication number
WO2018040673A1
WO2018040673A1 PCT/CN2017/088695 CN2017088695W WO2018040673A1 WO 2018040673 A1 WO2018040673 A1 WO 2018040673A1 CN 2017088695 W CN2017088695 W CN 2017088695W WO 2018040673 A1 WO2018040673 A1 WO 2018040673A1
Authority
WO
WIPO (PCT)
Prior art keywords
forwarding
low
latency
forwarding device
mode
Prior art date
Application number
PCT/CN2017/088695
Other languages
English (en)
French (fr)
Inventor
江元龙
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP17844979.9A priority Critical patent/EP3468123B1/en
Priority to EP21206178.2A priority patent/EP4030732A1/en
Publication of WO2018040673A1 publication Critical patent/WO2018040673A1/zh
Priority to US16/285,913 priority patent/US10972398B2/en
Priority to US17/222,201 priority patent/US11616729B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/50—Routing or path finding of packets in data switching networks using label swapping, e.g. multi-protocol label switch [MPLS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/12—Network monitoring probes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/17—Interaction among intermediate nodes, e.g. hop by hop
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2483—Traffic characterised by specific attributes, e.g. priority or QoS involving identification of individual flows
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/72—Admission control; Resource allocation using reservation actions during connection setup
    • H04L47/724—Admission control; Resource allocation using reservation actions during connection setup at intermediate nodes, e.g. resource reservation protocol [RSVP]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/82—Miscellaneous aspects
    • H04L47/825—Involving tunnels, e.g. MPLS

Definitions

  • the present application relates to the field of communications, and in particular, to a method and apparatus for processing a low latency traffic flow.
  • a low-latency service flow is a service flow that requires a transmission delay to be within a preset threshold, and its preset threshold is usually low, such as milliseconds.
  • each Ethernet port of an Ethernet network device has eight queues, and each queue stores the packets to be sent in the order of received packets.
  • Each queue corresponds to a gate for controlling message transmission, that is, each Ethernet port is also provided with 8 gates.
  • Each Ethernet port is configured with a gated list.
  • the gating list includes a plurality of entries, each of which includes eight gating values.
  • the eight ports included in the Ethernet port can be turned on or off according to the eight gate values included in each entry. In the case that the door is opened, the message to be sent in the queue corresponding to the opened door is sent.
  • Using a gated list for low-latency traffic processing requires configuring a gated list for each Ethernet port, which is complex and inflexible.
  • the embodiments of the present application provide a method and apparatus for processing a low-latency service flow, which helps reduce management complexity and control flexibility.
  • a method for processing a low latency traffic flow comprising:
  • the first forwarding device After the first forwarding device determines that the received first data packet belongs to the first service flow, the first forwarding device obtains a low-latency identifier corresponding to the first service flow, where the first forwarding device is a device that is a network entry;
  • the first forwarding device obtains a second data packet according to the first data packet and the low-latency identifier, where the second data packet includes the first data packet and the low-latency identifier.
  • the low-latency identifier is used to indicate that the forwarding device that receives the first service flow forwards the first service flow in a low-latency forwarding mode, where the low-latency forwarding mode is to implement the first service flow under dynamic control. Fast forwarding mode;
  • the first forwarding device sends the second data packet to the second forwarding device in the low-latency forwarding mode.
  • the low-latency service flow is a service flow that needs to be transmitted within a preset duration.
  • the preset duration is usually low, such as milliseconds, and may be 10 milliseconds or 5 milliseconds.
  • the transmission delay is a transmission delay required by the end to end in the network.
  • the low latency traffic flow may include one or more IP packets, or the low latency traffic flow may include one or more Ethernet frames.
  • the low-latency forwarding mode is a mode in which configuration is implemented by an instruction or information interaction, that is, dynamic control controls a state of the low-latency forwarding mode by an instruction or information interaction.
  • the low-latency forwarding mode is a mode that helps ensure that the low-latency service flow can be preferentially processed and fast forwarded.
  • the slot-gating function is used to preferentially process the low-latency service flow in the queue corresponding to the gate.
  • the time slot gating function is a function of controlling the door to be turned on or off within a preset time period.
  • the low-latency identifier is further used to identify a data packet that belongs to the first service flow.
  • the low-latency identifier may be used to identify that the second data packet belongs to the first service flow.
  • the first forwarding device inserts a low-latency identifier corresponding to the first service flow in the data packet that belongs to the first service flow, for example, in the first service flow.
  • a low-latency identifier corresponding to the first service flow is inserted into the first data packet to obtain a second data packet.
  • the first forwarding device forwards the second data packet in the low-latency forwarding mode, which helps accelerate the forwarding of data packets belonging to the low-latency service flow.
  • the first forwarding device can forward the first service flow according to the state of the low-latency forwarding mode, and does not need to configure a gating list on each port of the first forwarding device, which helps reduce management. The complexity and flexibility of control.
  • the first forwarding device may add the low-latency identifier to one or more data packets included in the first service flow, that is, one or more data packets included in the first service flow. After the first forwarding device processes, the same low delay identifier may be included. In this way, the device configuration on the forwarding path for forwarding the first service flow is simplified, which helps to quickly identify and process data packets belonging to the same service flow.
  • the first data packet may be the first data packet of the first service flow, or the last data packet of the first service flow, or the first service flow except the Any one of the first data message and the last data message.
  • the method further includes: the first forwarding device receiving the first data packet; A forwarding device determines whether the first data packet belongs to the first service flow; and the first forwarding device determines, according to the determination result, that the first data packet belongs to the first service flow.
  • the determining, by the first forwarding device, whether the first data packet belongs to the first service flow includes: the first forwarding device according to the port that receives the first data packet or the first data
  • the information carried in the packet determines whether the first data packet belongs to the first service flow.
  • the first forwarding device may store a feature information table of the first service flow, where the feature information table may include feature information for determining that the data packet belongs to the first service flow.
  • the feature information includes a Media Access Control (MAC) address, an Internet Protocol (IP) address, an application layer port number, a virtual local area network (VLAN) information, and a virtual extended local area network (Virtual Extensible).
  • MAC Media Access Control
  • IP Internet Protocol
  • VLAN virtual local area network
  • VPN virtual extended local area network
  • LAN, VXLAN Virtual Extensible
  • the first forwarding device may query the feature information table by using the information carried in the first data packet or the port receiving the first data packet, if the first datagram exists in the feature information table. The first forwarding device determines the first datagram if the information carries the same feature information, or if there is a physical port in the feature information table that matches the port that receives the first data packet.
  • the text belongs to the first service flow.
  • the first service flow belongs to a low-latency service flow.
  • the first data packet received by the first forwarding device may be the first data packet from the user that is forwarded by a user edge device (CE).
  • the first forwarding device is a device that is a network entry.
  • the first forwarding device may be an ingress operator edge (PE) device.
  • the second forwarding device may be a device on a forwarding path, for example, the second forwarding device may be a transit (provider, P) device.
  • the second forwarding device may be a device that is a network egress, for example, the second forwarding device may be an egress PE device.
  • the first forwarding device and the second forwarding device may be in the same forwarding path, and the forwarding path is a path for forwarding the first service flow.
  • the method before the sending, by the first forwarding device, the second data packet to the second forwarding device in the low-latency forwarding mode, the method further includes: the first forwarding device to the first The second forwarding device sends control information, and the control information is used to control the state of the low-latency forwarding mode.
  • the control information may correspond to the first service flow; and the low-latency forwarding mode is a forwarding mode corresponding to the first service flow.
  • the controlling the state of the low-latency forwarding mode includes controlling the opening and closing of the low-latency forwarding mode.
  • the control information includes a start time and an end time of the low-latency forwarding mode, or the control information includes a start time and a running time of the low-latency forwarding mode.
  • the controlling the state of the low-latency forwarding mode comprises controlling the opening of the low-latency forwarding mode.
  • the control information includes an open identifier, and the open identifier is used to identify that the low-latency forwarding mode is enabled. In this way, the first forwarding device sends the control message to the second forwarding device before processing the first service flow, omitting the configuration operation of the management plane, and implementing the low-latency service flow more flexibly. Processing.
  • the method further includes The first forwarding device sends a shutdown identifier to the second forwarding device, where the shutdown identifier is used to identify that the low-latency forwarding mode is disabled.
  • control information may be carried in a Resource Reservation Protocol (RSVP) message or a generic associated channel header (G-ACH) channel message.
  • RVP Resource Reservation Protocol
  • G-ACH generic associated channel header
  • control information may be carried in a type-length-value (TLV) included in the RSVP message or the G-ACH channel message.
  • TLV type-length-value
  • the method further includes: the first forwarding device to the The second forwarding device sends a detection packet, where the detection packet includes a first delay value and a second delay value, where the first delay value is a maximum duration of delay allowed by the forwarding device on the forwarding path, The second delay value is a delay duration generated by the first forwarding device.
  • the detection packet is used to obtain a transmission delay generated by the forwarding device on the forwarding path of the first service flow.
  • the sending, by the first forwarding device, the detection packet to the second forwarding device may be performed after the first forwarding device sends the control information to the second forwarding device.
  • the second forwarding device belongs to a forwarding path for forwarding the first service flow.
  • the first forwarding device sends the detection packet after the start time of the low-latency forwarding mode is set on the forwarding device included in the forwarding path, and the forwarding device in the low-delay forwarding mode state can be set.
  • the detection of the transmission delay helps to locate the forwarding device with a large transmission delay by using the detection result, thereby further reducing the end-to-end transmission delay.
  • control information sent by the first forwarding device to the second forwarding device further includes a bandwidth requirement of the first service flow.
  • the first forwarding device may send the control information, so that the second forwarding device allocates a corresponding second forwarding resource according to the bandwidth requirement of the first service flow, and does not need to send the forwarding resource again.
  • Resource allocation message may be used to send the control information, so that the second forwarding device allocates a corresponding second forwarding resource according to the bandwidth requirement of the first service flow, and does not need to send the forwarding resource again.
  • the low latency identifier may be a low latency tag.
  • the general multiprotocol label switching (MPLS) label for forwarding is extended to have a function of low delay identification and a forwarding function.
  • MPLS general multiprotocol label switching
  • a method for processing a low latency traffic flow comprising:
  • the second forwarding device receives the second data packet from the first forwarding device, where the second data packet includes a low-latency identifier, and the low-latency identifier is used to indicate that the forwarding device that receives the first service flow is forwarded at a low delay. Forwarding the first service flow in a mode, where the low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow, and the second data packet belongs to the first service flow;
  • the second forwarding device processes the second data packet in a low-latency forwarding mode according to the low-latency identifier.
  • the second forwarding device may determine whether to send the received data packet in the low delay mode according to whether the received data packet includes a low delay identifier, without performing the management plane. Complex configuration operations, and low latency mode is more flexible.
  • the second data packet further includes the first data packet. If the second forwarding device is an egress PE device, the second forwarding device performs the low delay forwarding mode according to the delay identifier. Processing the second data packet includes: deleting, by the second forwarding device, the low-latency identifier in the second data packet to obtain the first data packet; and the second forwarding device according to the delay identifier And transmitting, in the low delay mode, the first data packet to a CE device that is in communication with the second forwarding device.
  • the second forwarding device processes the second data packet in the low-latency forwarding mode according to the delay identifier, where the second forwarding device is configured according to the second forwarding device.
  • the delay flag, in the low delay mode, to the third The forwarding device sends the second data packet.
  • the third forwarding device is a next hop of the second forwarding device along the first direction on the forwarding path.
  • the forwarding path is a path for forwarding the first service flow, and the first direction is a direction from the ingress PE to the egress PE.
  • the third forwarding device is another transit P device or an egress PE device.
  • the method further includes: the second forwarding device receiving control information from the first forwarding device, The control information is used to control a state of the low-latency forwarding mode; the second forwarding device dynamically controls a state of the low-latency forwarding mode according to the control information.
  • control information includes a start time and an end time of the low-latency forwarding mode
  • the second forwarding device dynamically controls, according to the control information, a state of the low-latency forwarding mode, where the second The forwarding device runs the low-latency forwarding mode according to the start time and the end time of the low-latency forwarding mode.
  • control information includes a start time and a running time of the low-latency forwarding mode
  • the second forwarding device dynamically controls, according to the control information, a state of the low-latency forwarding mode, where the second The forwarding device runs the low-latency forwarding mode according to a start time of the low-latency forwarding mode and the running time.
  • control information includes an open identifier, where the open identifier is used to identify that the low-latency forwarding mode is enabled, and the second forwarding device sends the first identifier according to the low-latency identifier in a low-latency forwarding mode.
  • the method further includes: the second forwarding device receives the shutdown identifier sent by the first forwarding device, where the shutdown identifier is used to identify that the low-latency forwarding mode is disabled; The device stops sending the data packet of the first service flow in the low-latency forwarding mode according to the shutdown identifier.
  • the data packet that stops the sending of the first service flow in the low-latency forwarding mode may not forward the data packet of the first service flow.
  • the sending of the data packet of the first service flow in the low-latency forwarding mode may be performed by sending the data packet of the first service flow in a normal manner, that is, the first processing is not performed preferentially.
  • control information is sent is the same as the first aspect, and details are not described herein again.
  • the method further includes: the second forwarding device receives the first detection packet from the first forwarding device, where the first detection packet includes a first delay value and a second delay value.
  • the first detection packet is used to obtain a transmission delay generated by the forwarding device on the forwarding path, where the first delay value is a maximum duration of delay allowed by the forwarding device on the forwarding path, and the second The delay value is the delay time generated by the first forwarding device; the second forwarding device obtains a third delay value, and the third delay value is that the second forwarding device receives the first detection report.
  • the second detection device obtains a second detection message according to the first detection message and the third delay value, where the second detection device sends a second detection message.
  • the text includes the first detection message and the third delay value.
  • the second forwarding device in the embodiment of the present application may generate the first detection packet according to the received detection packet from the first forwarding device, such as the first detection packet, and the transmission delay of the analog data packet.
  • the transmission delay is added to the first detection message to obtain a second detection message.
  • the device that receives the second detection packet can use the information and/or parameters carried in the second detection packet to locate a forwarding device with a large transmission delay, which helps to further reduce end-to-end transmission. Delay.
  • the second forwarding device is a transit P device
  • the method further includes: the second forwarding device sends the second detection packet to the third forwarding device, where the third forwarding device is Forwarding the next hop of the second forwarding device on the path.
  • a first forwarding device where the first forwarding device includes:
  • a processing unit configured to determine that the received first data packet belongs to the first service flow, and obtain a low-latency identifier corresponding to the first service flow, where the first forwarding device is a device that is a network entry;
  • An obtaining unit configured to obtain a second data packet according to the first data packet and the low delay identifier, where the second data packet includes the first data packet and the low delay identifier, where The low-latency identifier is used to indicate that the forwarding device that receives the first service flow forwards the first service flow in a low-latency forwarding mode, where the low-latency forwarding mode is to implement the first service flow under dynamic control.
  • Fast forwarding mode
  • the first sending unit is configured to send the second data packet to the second forwarding device in the low-latency forwarding mode.
  • the first forwarding device further includes:
  • a second sending unit configured to send, to the second forwarding device, control information, where the control information is used to control a state of the low-latency forwarding mode.
  • the second sending unit may send the the second forwarding device to the second forwarding device before the first sending unit sends the second data packet to the second forwarding device in the low-latency forwarding mode.
  • Control information if the first data packet is the first data packet of the first service flow, the first forwarding device sends the second data packet including the first data packet, The control information for controlling the state of the low-latency forwarding mode is sent to the second forwarding device, so that the configuration of the low-latency forwarding mode is more flexible.
  • control information includes a start time and an end time of the low-latency forwarding mode, or the control information includes a start time and a running time of the low-latency forwarding mode.
  • control information includes an open identifier, where the open identifier is used to identify that the low-latency forwarding mode is enabled, and the first forwarding device further includes:
  • a third sending unit configured to send a shutdown identifier to the second forwarding device, where the shutdown identifier is used to identify that the low-latency forwarding mode is disabled.
  • the first forwarding device further includes:
  • a fourth sending unit configured to send a detection packet to the second forwarding device, where the detection packet includes a first delay value and a second delay value, where the first delay value is forwarding on the forwarding path The maximum duration of delay allowed by the device, where the second delay value is a delay duration generated by the first forwarding device.
  • the detection packet is used to obtain a transmission delay generated by the forwarding device on the forwarding path.
  • the forwarding path includes the first forwarding device and the second forwarding device, and the forwarding path is a path for forwarding the first service flow.
  • the detecting packet further includes a time when the first forwarding device sends the detection packet.
  • the forwarding device that is the network egress can be carried according to the time when the first forwarding device sends the detection packet, the time when the forwarding device that is the network egress receives the detection packet, and the detection packet carries.
  • the delay value generated by the forwarding device determines the delay value generated by the forwarding device and the delay value generated by the physical link on the forwarding path, which helps locate a forwarding device and/or a physical link with a large delay value.
  • the first forwarding device provided by any one of the foregoing third aspect or the third aspect may adopt the method provided by the first aspect or any one of the possible implementation manners of the first aspect.
  • a second forwarding device where the second forwarding device includes:
  • a first receiving unit configured to receive a second data packet from the first forwarding device, where the second data packet includes a low-latency identifier, where the low-latency identifier is used to indicate that the forwarding device that receives the first service flow is Forwarding the first service flow in a low-latency forwarding mode, where the low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow, and the second data packet belongs to the first service flow ;
  • a first sending unit configured to process the second data packet in the low-latency forwarding mode according to the low-latency identifier.
  • the second forwarding device further includes:
  • a second receiving unit configured to receive control information from the first forwarding device, where the control information is used to control a state of the low-latency forwarding mode
  • control unit configured to dynamically control a state of the low-latency forwarding mode according to the control information.
  • the second receiving unit may receive the control information before the first receiving unit receives the second data packet, so that the control unit may receive the first receiving unit.
  • dynamic control of the state of the low-latency forwarding mode is completed, such as enabling the low-latency forwarding mode.
  • control information includes a start time and an end time of the low-latency forwarding mode, where the control unit is specifically used for root
  • the low-latency forwarding mode is operated according to the start time and the end time of the low-latency forwarding mode.
  • control information includes a start time and a running time of the low-latency forwarding mode, where the control unit is specifically configured to run the low delay according to a start time of the low-latency forwarding mode and the running duration Forward mode.
  • control information includes an open identifier, where the open identifier is used to identify that the low-latency forwarding mode is enabled, and the second forwarding device further includes:
  • a third receiving unit configured to receive a shutdown identifier sent by the first forwarding device, where the shutdown identifier is used to identify that the low-latency forwarding mode is disabled;
  • the control unit is further configured to stop sending the data packet of the first service flow in the low-latency forwarding mode according to the shutdown identifier.
  • the second forwarding device further includes:
  • a fourth receiving unit configured to receive a first detection packet from the first forwarding device, where the first detection packet includes a first delay value and a second delay value, where the first detection packet is used by the first detection packet Obtaining a transmission delay generated by the forwarding device on the forwarding path, where the first delay value is a maximum duration of delay allowed by the forwarding device on the forwarding path, and the second delay value is the first forwarding device The length of delay generated;
  • a first obtaining unit configured to obtain a third delay value, where the third delay value is a length of time that the second forwarding device receives the first detection packet to send the first detection packet;
  • a second obtaining unit configured to obtain a second detection packet according to the first detection packet and the third delay value, where the second detection packet includes the first detection packet and the first Three time delay value.
  • the second forwarding device is an intermediate forwarding device, where the first sending unit is configured to send the second detection packet to the third forwarding device, where the third forwarding device is on the forwarding path.
  • the second forwarding device provided by any one of the foregoing fourth aspect or the fourth aspect may adopt the method provided by the second aspect or any one of the possible implementation manners of the second aspect.
  • a first forwarding device comprising: a processor, a memory, and a communication interface.
  • the processor, the memory, and the communication interface are connected by a communication bus.
  • the memory is used to store programs.
  • the processor performs the method provided by any one of the first aspect or the first aspect of the first aspect, in accordance with executable instructions included in a program read from the memory.
  • the first forwarding device provided by the sixth aspect may be the first forwarding device provided by the third aspect.
  • a second forwarding device comprising: a processor, a memory, and a communication interface.
  • the processor, the memory, and the communication interface are connected by a communication bus.
  • the memory is used to store programs.
  • the processor performs the method provided by any one of the possible implementations of the second aspect or the second aspect, according to the executable instructions included in the program read from the memory.
  • the seventh aspect provides a system for processing a low-latency service flow, where the system includes the first forwarding device provided by any one of the foregoing third aspect or the third aspect, and the foregoing fourth aspect or The second forwarding device provided by any one of the possible implementation manners of the fourth aspect; or the system includes the first forwarding device provided by any one of the foregoing fifth aspect or the fifth aspect, and the foregoing sixth aspect or A second forwarding device provided by any one of the possible implementation manners of the sixth aspect.
  • Figure 1 is a schematic diagram of a port of a bridge.
  • FIG. 2 is a schematic diagram of a network scenario according to an embodiment of the present disclosure.
  • FIG. 3 is a flowchart of a method for configuring a low-latency service forwarding mode according to Embodiment 1 of the present application.
  • FIG. 4 is a flowchart of a method for processing a low-latency service flow according to Embodiment 2 of the present application.
  • FIG. 5(a) is a schematic diagram of a first G-ACH channel message according to an embodiment of the present application.
  • FIG. 5(b) is a schematic diagram of a second data packet according to an embodiment of the present application.
  • FIG. 5(c) is a schematic diagram of a second data packet according to an embodiment of the present application.
  • FIG. 6 is a flowchart of a method for processing a low-latency service flow according to Embodiment 3 of the present application.
  • FIG. 7 is a flowchart of a method for detecting a transmission delay of a forwarding path according to Embodiment 4 of the present application.
  • FIG. 8 is a schematic diagram of a first forwarding device according to Embodiment 5 of the present application.
  • FIG. 9 is a schematic diagram of a second forwarding device according to Embodiment 6 of the present application.
  • FIG. 10 is a schematic diagram of a first forwarding device according to Embodiment 7 of the present application.
  • FIG. 11 is a schematic diagram of a second forwarding device according to Embodiment 8 of the present application.
  • FIG. 12 is a flowchart of a method for establishing a forwarding path according to Embodiment 10 of the present application.
  • FIG. 13 is a schematic diagram of a network scenario for establishing a forwarding path according to an embodiment of the present disclosure.
  • Figure 1 is a schematic diagram of a port of a bridge.
  • a port of the bridge includes 8 queues and 8 gates.
  • the eight queues include queue 0, queue 1, queue 2, queue 3, queue 4, queue 5, queue 6, and queue 7.
  • the eight doors include a door 0, a door 1, a door 2, a door 3, a door 4, a door 5, a door 6, and a door 7.
  • Each queue corresponds to a send selection algorithm.
  • the sending selection algorithm is used to calculate the priority of the traffic to be sent in the queue and output the traffic to the corresponding gate according to the priority.
  • the port of the bridge is set with a gate list, and the gate list includes 80 gate entries, such as T00, T01, T02...T78 and T79.
  • Each gating entry is used to hold gating signals within each designated time slot.
  • Each bit of the gating signal is used to control the corresponding door to be opened or closed.
  • C in Fig. 1 indicates off, and O in Fig. 1 indicates on.
  • the gating signal included in T04 is OCOOCOOO, that is, the door 7 is opened, the door 6 is closed, the door 5 is opened, and the door 4 is opened.
  • Door 3 is closed, door 2 is open, door 1 is open and door 0 is open.
  • the high priority traffic in queue 7 is output to gate 7.
  • the gate 7 Since the gate 7 is in the on state according to the corresponding gating signal of T04, the high priority traffic flow in the queue 7 is output. After the high priority traffic in queue 6 is selected by the sending selection algorithm, the high priority traffic in queue 6 is output to gate 6. Gate 6 is in the off state according to the corresponding gating signal of T04, and the high priority traffic in queue 6 stops outputting. A port of the bridge is the same as the queue 6 and the queue 7 in the queue 5, the queue 4, the queue 1 and the queue 0, and is not described here.
  • Ports of bridges associated with low-latency services require a gated list to be configured through the management plane, increasing management complexity. Since the number of gating entries included in the gating list is limited, it may be necessary to perform corresponding configuration according to the service flow before all the gating entries are used, which is less flexible.
  • the present application proposes a method that contributes to reducing the complexity of management and improving the flexibility of control.
  • the first forwarding device that is the network entry determines the received first data packet, and after determining that the first data packet belongs to the first service flow, the first forwarding device is configured according to the first And obtaining a second data packet by using a data packet and a low delay identifier corresponding to the first service flow.
  • the first forwarding device sends the second data packet to the second forwarding device in the low-latency forwarding mode.
  • the second forwarding device sends the second data packet in the low-latency forwarding mode according to the low-latency identifier included in the second data packet.
  • the low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow under dynamic control.
  • the low-latency service flow in the embodiment of the present application is a service flow that requires the transmission delay to be within a preset duration.
  • the preset duration is usually longer than Low, such as milliseconds, can be 5 milliseconds or 10 milliseconds.
  • the transmission delay is a delay generated in an end-to-end transmission process in the network.
  • the end-to-end transmission in the network refers to the transmission between the device as the network entry to the device as the network egress, or the end-to-end transmission in the network refers to the source address of the message. Transfer to the destination address of the message.
  • the low latency traffic flow may include one or more IP packets, or the low latency traffic flow may include one or more Ethernet frames.
  • the low-latency forwarding mode in the embodiment of the present application is a mode for implementing fast forwarding of the first service flow under dynamic control, which helps ensure that low-latency service flows can be preferentially processed, for example, using a time slot gating function for corresponding
  • the low latency traffic in the queue is prioritized.
  • the time slot gating function is a function of controlling a door to be turned on or off within a preset time period, and the door can be used to transmit a service flow in a corresponding queue thereof.
  • the priority processing includes priority scheduling and priority forwarding.
  • FIG. 2 is a schematic diagram of a network scenario according to an embodiment of the present disclosure.
  • CE1, CE2, CE3, and CE4 are devices that can communicate with users.
  • PE1 can communicate with CE1, CE2 and P1.
  • P1 can communicate with PE2.
  • PE2 can communicate with CE3 and CE4.
  • CE1 sends the first service flow to CE3.
  • the solid line in Figure 2 represents the path used to forward the first traffic flow.
  • the path for forwarding the first service flow is a first forwarding path.
  • CE2 sends a second service flow to CE4.
  • the dashed line in Figure 2 represents the path used to forward the second traffic flow.
  • the path for forwarding the second service flow is a second forwarding path.
  • the first service flow and the second service flow are both low latency traffic flows.
  • PE1 may be an ingress node of the first forwarding path and an ingress node of the second forwarding path.
  • PE2 may be an egress node of the first forwarding path and an egress node of the second forwarding path.
  • the devices PE1, PE2, and P1 in Figure 2 are devices in the carrier network.
  • any of the CEs in Figure 2 may be a broadband access client, an enterprise egress gateway, or an output gateway of a data center (DC).
  • Any of the PEs in Figure 2 may be routers or Packet Transport Network (PTN) devices.
  • Any P in Figure 2 can be a router or a PTN device.
  • the link between any of the CEs in FIG. 2 and the adjacent communicable PEs may be an Ethernet link, a passive optical network (PON) link, or an x digital subscriber line (x digital subscriber line, xDSL) link.
  • PON passive optical network
  • xDSL x digital subscriber line
  • the link between CE1 and PE1, the link between CE2 and PE1, the link between CE3 and PE2, and the link between CE4 and PE2 may be Ethernet links, PON links, or xDSL links. The combination of different links is illustrated.
  • FIG. 3 is a flowchart of a method for configuring a low-latency service forwarding mode according to Embodiment 1 of the present application.
  • a method for processing a low-latency service flow is described from the perspective of configuring a low-latency forwarding mode on a device on a forwarding path.
  • All devices in the carrier network of Figure 2 need to be pre-deployed with time synchronization protocols, such as the 1588 Clock Synchronization Protocol of the Institute of Electrical and Electronics Engineers (IEEE), IEEE 1588v2, which makes all the carriers in the network.
  • the device has time to synchronize across the network.
  • the forwarding device in the carrier network may perform time synchronization by using the foregoing clock synchronization protocol before configuring the low-latency forwarding mode, and may also perform frequency synchronization by using the above-mentioned frequency synchronization protocol.
  • the method in this application does not describe the method of clock synchronization and/or frequency synchronization of all devices in the carrier network.
  • the ingress node in the embodiment of the present application is an entry for forwarding a forwarding path of a service flow in the operator network, such as PE1.
  • the egress node is an egress of the forwarding path for forwarding a service flow in the carrier network, such as PE2.
  • Embodiment 1 of the present application includes a configuration process of a first low-latency forwarding mode and a configuration process of a second low-latency forwarding mode.
  • a method for configuring a low-latency service forwarding mode according to Embodiment 1 of the present application is described below with reference to FIG. 2 and FIG.
  • PE1 sends first control information to P1.
  • the first control information is a state for controlling the first low-latency forwarding mode.
  • the first low-latency forwarding mode is a low-latency forwarding mode adopted by the device for processing the first service flow, that is, the first low-latency forwarding mode corresponds to the first service flow.
  • the first low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow under dynamic control.
  • the device for processing the first service flow may be a forwarding device in the carrier network of FIG. 2 for processing the first service flow.
  • the first low-latency forwarding mode is set on PE1, P1, and PE2.
  • the state of the first low-latency forwarding mode includes a start of the first low-latency forwarding mode and an end of the first low-latency forwarding mode. Or the state of the first low-latency forwarding mode includes a start of the first low-latency forwarding mode.
  • the first control information includes a start time of the first low-latency forwarding mode and an end time of the first low-latency forwarding mode.
  • the first control information includes a start time of the first low-latency forwarding mode and a running time of the first low-latency forwarding mode.
  • the first control information includes a first open identifier. The first open identifier is used to indicate that the device that obtains the first open identifier turns on the first low-latency forwarding mode.
  • the start time of the first low-latency forwarding mode may be represented as T1.
  • the end time of the first low-latency forwarding mode may be represented as T2.
  • the runtime of the first low latency forwarding mode may be denoted as t.
  • the manner of acquiring T1 and T2 may be any one of mode one to mode three.
  • the CE1 obtains T1 and T2 of the authenticated user (the user who communicates with CE1 and passes the authentication), and PE1 obtains T1 and T2 from the authentication device.
  • T1 can be set as the moment when the user passes the authentication.
  • the time in the embodiment of the present application refers to a certain time point.
  • the user purchases the first service through the page provided by the operator.
  • the PE1 can obtain the T1 and the T2 from the server provided by the operator.
  • PE1 obtains T1 and T2 from the user (the user who communicates with CE1).
  • the method for the PE1 to obtain the T1 and the T2 from the user may be that the user actively reports T1 and T2 to the PE1.
  • the first service flow is a data flow corresponding to the first service.
  • the manners of obtaining the T1 and the T2 are the same as those of the foregoing T1 and T2, and are not described here.
  • the PE1 sends the first control information in the process of establishing the first forwarding path, where the first forwarding path is a path for forwarding the first service flow.
  • the first forwarding path is a path for forwarding the first service flow.
  • P1 is a device on the first forwarding path.
  • the first control information may be carried in a Resource Reservation Protocol (RSVP) message or a generic associated channel header (G-ACH) channel message.
  • RSVP Resource Reservation Protocol
  • G-ACH generic associated channel header
  • the RSVP message carrying the first control information may be extended.
  • the RSVP message may be extended in the RSVP message, for example, a type-length-value (TLV) is added.
  • TLV carried by the low delay field may be used to carry the first control information.
  • the TLV carried by the low-latency field may also be used to carry a first control flag (in English, a flag).
  • the first control flag is used to identify the opening or closing of the sub-forwarding mode.
  • the sub-forwarding mode may be a normal preemptive forwarding or a time slot scheduling.
  • the first G-ACH channel message may be extended.
  • an associated channel header (ACH) TLV is added to the first G-ACH channel message, and the ACH TLV can be used to carry the first control information, as shown in FIG. 5(a).
  • a generic associated channel label (GAL), a PW label, and a first label may be sequentially encapsulated in the outer layer of the first G-ACH channel message.
  • the GAL is used to indicate the presence of a G-ACH control channel.
  • the PW label is optional. For example, for some Layer 2 services, a PW label needs to be added. For a Layer 3 service, there is no need to add a PW label.
  • the first G-ACH channel message used to carry the first control information in the embodiment of the present application may adopt the structure in FIG. 5(a), which is not illustrated one by one in the following embodiments.
  • the first low-latency forwarding mode further includes
  • the ACH TLV may also be used to carry the first control flag.
  • the first control flag is used to identify the opening or closing of the sub-forwarding mode.
  • the sub-forwarding mode may be a normal preemption forwarding or timescale scheduling.
  • PE1 sends second control information to PE2.
  • the second control information is a state for controlling the second low-latency forwarding mode.
  • the second low-latency forwarding mode is a low-latency forwarding mode adopted by the device for processing the second service flow, that is, the second low-latency forwarding mode corresponds to the second service flow.
  • the second low-latency forwarding mode is a mode for implementing fast forwarding of the second service flow under dynamic control.
  • the device for processing the second service flow may be a forwarding device in the carrier network of FIG. 2 for processing the second service flow.
  • the second low-latency forwarding mode is also set on PE1 and PE2.
  • the state of the second low latency forwarding mode includes the beginning and the end of the second low latency forwarding mode.
  • the state of the second low-latency forwarding mode includes a start of the second low-latency forwarding mode.
  • the beginning of the second low latency forwarding mode refers to the opening of the second low latency forwarding mode.
  • the end of the second low latency forwarding mode refers to the end of the second low latency forwarding mode.
  • the second control information includes a start time of the second low-latency forwarding mode and an end time of the second low-latency forwarding mode.
  • the second control information includes a start time of the second low-latency forwarding mode and a running time of the second low-latency forwarding mode.
  • the second control information includes a second open identifier. The second open identifier is used to indicate that the device that obtains the second open identifier turns on the second low-latency forwarding mode.
  • PE1 sends the second control information in the process of establishing a second forwarding path.
  • the second forwarding path is configured to transmit the second service flow.
  • PE2 is a device on the second forwarding path, and is also a device on the second forwarding path as an egress node.
  • the sending manner of the second control information is the same as the sending manner of the first control information, and details are not described herein again.
  • the RSVP message may be extended.
  • the manner of extending the RSVP message to carry the second control information is the same as the manner of extending the RSVP message to carry the first control information, and details are not described herein.
  • the second G-ACH channel message carrying the second control information may be extended, for example, the ACH TLV included in the second G-ACH message may be used to carry the Two control information.
  • the manner of extending the second G-ACH channel message to carry the second control information is the same as the manner of expanding the first G-ACH channel message, and details are not described herein.
  • P1 is configured according to the first control information.
  • P1 configures a state of the first low-latency forwarding mode according to the first control information.
  • P1 may generate an entry corresponding to the first service flow according to the first control information. If the first control information includes a start time of the first low-latency forwarding mode and an end time of the first low-latency forwarding mode, the entry corresponding to the first service flow includes: a first low latency identifier, a start time of the first low latency forwarding mode, and an end time of the first low latency forwarding mode.
  • the entry corresponding to the first service flow includes: a first low latency identifier, a start time of the first low latency forwarding mode, and a runtime of the first low latency forwarding mode.
  • the first low latency identifier may be derived from the first control information, or obtained by static configuration, or by other messages from PE1.
  • P1 may generate an entry corresponding to the first service according to the first control information.
  • the entry corresponding to the first service flow further includes the identifier of the first service flow
  • the entry corresponding to the first service further includes an identifier of the first service, that is, The identifier of the first service is replaced by the identifier of the first service flow that is included in the entry corresponding to the first service flow, to obtain the entry corresponding to the first service.
  • P1 sends the first control information to PE2.
  • P1 may send the first control information to the PE2 along the first forwarding path.
  • the execution order of 301, 302, and 304 needs to ensure that 301 is executed before 303 and 304, and the execution order of 302 is not limited.
  • 302 can Execute at the same time as 301, or 302 may be performed prior to 301. Or 302 may be performed concurrently with 303, or 302 may be performed after 303. Or 302 may be performed concurrently with 304, or 302 may be performed after 304. 303 may be performed after 304, or 303 may be performed concurrently with 304.
  • the PE2 is configured according to the first control information.
  • the PE2 configures the state of the first low-latency forwarding mode according to the first control information.
  • the configuration of the state of the first low-latency forwarding mode is the same as that of the configuration of the P1 in the 303, and is not described here.
  • the PE2 can obtain the entry corresponding to the first service flow or obtain the entry corresponding to the first service by using the method of P1 in 303.
  • the entry corresponding to the first service flow obtained by the PE2 may be referred to as a first entry.
  • the entry corresponding to the first service flow obtained by PE2 is the same as the entry corresponding to the first service flow in P1.
  • the entry corresponding to the first service obtained by the PE2 may be referred to as a first entry.
  • the entry corresponding to the first service obtained by PE2 is the same as the entry corresponding to the first service obtained by P1 in 303.
  • the PE2 is configured according to the second control information.
  • PE2 configures the state of the second low-latency forwarding mode according to the second control information.
  • the PE2 may generate a second entry according to the second control information, where the second entry is an entry corresponding to the second service flow. If the second control information includes a start time of the second low-latency forwarding mode and an end time of the second low-latency forwarding mode, the second entry includes: the second low-latency identifier, the first The start time of the two low-latency forwarding mode and the end time of the second low-latency forwarding mode.
  • the second entry includes: the second low-latency identifier, the The start time of the second low-latency forwarding mode and the running time of the second low-latency forwarding mode are described.
  • the second low latency identifier may be derived from the second control information, either by static configuration or by other messages from PE1.
  • the PE2 may generate an entry corresponding to the second service according to the second control information, that is, the entry corresponding to the second service is the second entry. And if the entry corresponding to the second service flow further includes the identifier of the second service flow, replacing the identifier of the second service flow with the identifier of the second service, to obtain the The entry corresponding to the second service.
  • 305 is performed after 304 and 306 is performed after 302.
  • the order of 305 and 306 is not limited, for example, 305 can be executed simultaneously with 306, or 305 can be executed after 306.
  • the first control information may further include a first low-latency (low latency) identifier.
  • the first low latency identifier corresponds to the first traffic flow.
  • the first low-latency identifier is used to indicate that the forwarding device that receives the first service flow forwards the first service flow in the first low-latency forwarding mode.
  • the specific form of the first low-latency identifier is not limited in this embodiment of the present application.
  • the PE1 can send the first low-latency identifier by using the message carrying the first control information, which saves the number of packets sent in the network, and helps save network resources.
  • the PE1 may further send the message carrying the first low-latency identifier along the first forwarding path before, after, or at the same time of sending the first control information.
  • the first control information is sent by the PE1 to the first control information, and the first control information further includes a bandwidth requirement of the first service flow and a delay value allowed by the first service flow.
  • the allowed delay value of the first service flow is a delay value allowed by a forwarding device on the first forwarding path.
  • the device that receives the bandwidth requirement of the first service flow and the delay value allowed by the first service flow may use the bandwidth requirement of the first service flow and the delay value allowed by the first service flow.
  • the first forwarding resource is configured to process the first service flow.
  • the PE1 may further send, according to the first forwarding path, a bandwidth requirement that carries the first service flow and a delay allowed by the first service flow before, after, or at the same time of sending the first control information. Value message.
  • the second control information may further include a second low-latency identifier.
  • the second low latency identifier corresponds to the second traffic flow.
  • the second low-latency identifier is used to indicate that the forwarding device that receives the second service flow forwards the second service flow in the second low-latency forwarding mode.
  • the specific form of the second low-latency identifier is not limited in this embodiment of the present application.
  • PE1 can send the second low-latency identifier by using a message carrying the second control information, thereby saving network. The number of packets sent in the middle helps save network resources.
  • the PE1 may further send the message carrying the second low-latency identifier along the second forwarding path before, after or simultaneously with sending the second control information.
  • the second control information is sent by the PE1 to the PE2 to the second control information.
  • the second control information further includes a bandwidth requirement of the second service flow and a delay value allowed by the second service flow.
  • the allowed delay value of the second service flow is a delay value allowed by a forwarding device on the second forwarding path.
  • the device that receives the bandwidth requirement of the second service flow and the delay value allowed by the second service flow may use the bandwidth requirement of the second service flow and the delay value allowed by the second service flow. Allocate the second forwarding resource.
  • the second forwarding resource is configured to process the second service flow.
  • the PE1 may further send, according to the second forwarding path, a bandwidth requirement that carries the second service flow and a time delay allowed by the second service flow before, after, or at the same time of sending the second control information. Value message.
  • the P1 may be configured according to the bandwidth requirement of the first service flow included in the first control information and the time allowed by the first service flow.
  • the delay value is allocated to forward the first forwarding resource of the first service flow.
  • P1 may allocate a snatch priority queue matching the first service flow to the first service flow, that is, allocate a leak bucket matching the first service flow for the first service flow (bucket in English) .
  • the leaky bucket matching the first service flow is a product of a bandwidth requirement of the first service flow and a delay value allowed by the first service flow. In this way, the actual delay generated by the first service flow at P1 is smaller than the delay value allowed by the first service flow.
  • P1 may also allocate a credit growth rate (in English, a credit rate) matching the first service flow.
  • the credit rate indicates a total number of bytes of data packets that can be sent within the allowed delay value of the first traffic flow, relative to the bucket that matches the first traffic flow.
  • the manner in which the PE2 allocates the first forwarding resource for forwarding the first service flow is the same as the allocation mode used by P1 in 303. No longer.
  • the PE2 may further allocate a credit rate matching the first service flow to the first service flow, and the specific allocation mode and the P1 in the 303.
  • the allocation method used is the same and will not be described here.
  • the PE2 is configured to forward the second forwarding resource of the second service flow, and the P1 allocation in the 303 is used to forward the first
  • the first forwarding resource of the service flow is in the same manner and will not be described here.
  • the leaky bucket matching the second service flow is a product of a bandwidth requirement of the second service flow and a delay value allowed by the second service flow, so as to help implement the second service flow.
  • the actual delay generated at PE2 is less than the delay allowed by the second traffic flow.
  • the PE2 may also allocate a credit rate for the second service flow to match the second service flow.
  • the credit rate indicates a total number of bytes of data packets that can be transmitted within a delay value allowed by the second traffic flow, relative to the bucket that matches the second traffic flow.
  • the method provided in Embodiment 1 of the present application further includes: PE1 performing configuration of the first low-latency forwarding mode according to the first control information. Before processing the first service flow, the PE1 performs configuration of the first low-latency forwarding mode according to the first control information.
  • the configuration method of the first low-latency forwarding mode of the PE1 is the same as that of the P1, and is not described here.
  • the method provided in Embodiment 1 of the present application further includes: PE1 performing configuration of the second low-latency forwarding mode according to the second control information. Before processing the second service flow, the PE1 performs configuration of the second low-latency forwarding mode according to the second control information.
  • the configuration method of the second low-latency forwarding mode of the PE1 is the same as that of the PE2 in the 306, and is not described here.
  • the first low-latency identifier in the embodiment of the present application is further used to indicate a data packet that belongs to the first service flow.
  • the second low latency identifier is further configured to indicate a data packet that belongs to the second service flow.
  • the configuration process of the first low-latency forwarding mode and the configuration process of the second low-latency forwarding mode are mutually independent processes.
  • the device that is the network entry such as the PE1 sends the first control information to the device on the first forwarding path, so as to receive the first control information, such as P1 or PE2. And configuring a state of the first low-latency forwarding mode according to the first control information to complete configuration of the first low-latency forwarding mode.
  • the PE1 may also send the second control information to the device on the second forwarding path, so that the device that receives the second control information, such as PE2, configures the state of the second low-latency forwarding mode according to the second control information, to The configuration of the second low-latency forwarding mode is completed.
  • the device as a network entry may dynamically control the low-latency forwarding mode of the device on the forwarding path by sending the first control information and/or the second control information.
  • the device on the first forwarding path and/or the second forwarding path does not need to configure a gating list corresponding to the service flow through the management plane, and does not need to perform maintenance and update operations on the gating list, thereby helping to reduce the management plane.
  • the carrier network is a Multiprotocol Label Switching (MPLS) network
  • the low-latency service flow is the first service flow
  • a method for processing the low-latency service flow is described.
  • the MPLS protocol runs on PE1, P1, and PE2 in Figure 2.
  • the first forwarding path for forwarding the first service flow is a label switched path (LSP).
  • LSP label switched path
  • the first low latency identifier corresponding to the first traffic flow is a first low latency label.
  • PE1, P1, and PE2 complete the configuration of the first low-latency forwarding mode before processing the first service flow.
  • Embodiment 1 The method provided in the first embodiment can be inserted into the second embodiment to form another embodiment.
  • the embodiment formed in the second embodiment will not be described.
  • a label corresponding to the first LSP may be configured on the PE1, the P1, and the PE2, and the label corresponding to the first LSP is used to forward the packet along the first LSP.
  • IETF RFC3209 The distribution process of the label corresponding to the first LSP is not described in the second embodiment of the present application.
  • a method for processing a low-latency service flow provided in Embodiment 2 of the present application will be described below with reference to FIG. 2 and FIG.
  • the PE1 receives the first data packet.
  • PE1 receives the first data packet through the first port.
  • the first data message may be from CE1.
  • the first data message includes a first MAC address and a first IP address.
  • the first MAC address is a MAC address of CE1.
  • the first IP address is an IP address of CE1.
  • the first data packet may further include a second MAC address and a second IP address, where the second MAC address is a MAC address of the CE3.
  • the second IP address is an IP address of CE3.
  • CE1 and CE3 can belong to the same virtual private network (VPN).
  • the first data message may be an IP packet or an Ethernet frame.
  • the method provided in the first embodiment of the present application further includes: the PE1 determining whether the first data packet belongs to the first service flow; and if the first data packet belongs to the first Traffic flow, PE1 executes 402. If the first data packet does not belong to the first service flow, the PE1 processes the first data packet into the to-be-forwarded queue, and processes the packet in the forwarding queue according to a normal scheduling method.
  • the normal scheduling method may be performed in the order of the enqueue.
  • the embodiment of the present application does not describe the usual scheduling methods one by one.
  • the manner in which the PE1 determines whether the first data packet belongs to the first service flow may adopt any one of the first mode and the fourth mode.
  • the first mode may include the first port, the first MAC address, the first IP address, the second MAC address, and the second IP address, and a service class (class of service, One or more of the port numbers of CoS), Traffic Class (TC), and Transmission Control Protocol (TCP).
  • the PE1 may determine, according to the multi-group, whether the first data packet belongs to the first service flow.
  • a feature information table of the first service flow may be stored on the PE1.
  • the feature information table may include the identifier of the first service flow and the feature information corresponding to the data packet belonging to the first service flow.
  • the feature information includes one or more of a MAC address, an IP address, an application layer port number, VLAN information, VXLAN information, and a physical layer port number.
  • the VLAN information may include a VLAN identifier (ID) and/or a priority.
  • the VXLAN information may include a VXLAN Network Identifier (VNI).
  • PE1 queries the characteristics of the first service flow by using the multi-group Information Sheet. If the feature information table of the first service flow has information that matches the multi-group, the PE1 determines that the first data packet belongs to the first service flow. The matching refers to the same information as the multi-group in the feature information table of the first service flow. For example, the multi-group includes N pieces of information, and the N is greater than or equal to 1 integer. If the information information of the first service flow has the same information as the N pieces of information, the PE1 may determine the The first data packet belongs to the first service flow.
  • the second mode the feature information table saved on the PE1 includes the identifier of the first service and the feature information corresponding to the data packet belonging to the first service.
  • the feature information corresponding to the data packet belonging to the first service may be the same as the feature information corresponding to the data packet belonging to the first service flow in the first mode.
  • the method for determining, by the PE1, whether the first data packet belongs to the first service flow is the same as the first manner, according to the feature information corresponding to the data packet that belongs to the first service. If the first data packet received by the PE1 includes the feature information corresponding to the data packet of the first service flow, the PE1 determines that the received first data packet belongs to the first service flow.
  • the third mode the first data packet further includes an identifier of the service flow to which the first data packet belongs. If the identifier of the service flow to which the first data packet belongs is the identifier of the first service flow, the PE1 determines that the first data packet belongs to the first service flow.
  • the fourth mode the first data packet further includes an identifier of the service to which the first data packet belongs. If the identifier of the service to which the first data packet belongs is the identifier of the first service, the PE1 determines that the first data packet belongs to the first service, that is, the first data packet belongs to the The first business flow.
  • the first service flow is a data flow of the first service.
  • the PE1 determines that the first data packet belongs to the first service flow.
  • the PE1 may obtain the first low delay label after determining that the first data packet belongs to the first service flow.
  • the manner in which the PE1 obtains the first low-latency label may be any one of the first mode and the third mode.
  • the PE1 may obtain the identifier of the first service flow after determining that the first data packet belongs to the first service flow according to the feature information table in the 401. Or the first data packet carries the identifier of the first service flow, and the PE1 obtains the identifier of the first service flow from the first data packet. A correspondence between the first low delay label and the identifier of the first service flow is configured on the PE1. The PE1 may obtain the first low delay label according to the correspondence and the identifier of the first service flow.
  • the PE1 may obtain the identifier of the first service after determining that the first data packet belongs to the first service flow. Or the first data packet carries the identifier of the first service, and the PE1 obtains the identifier of the first service from the first data packet. A correspondence between the first low delay label and the identifier of the first service is configured on the PE1. The PE1 may obtain the first low delay label according to the correspondence and the identifier of the first service.
  • Manner 3 If only the first low-latency label is configured on the PE1, and the PE1 is only used to forward the first service flow, the PE1 determines that the first data packet belongs to the first service flow, and directly Obtaining the first low latency tag.
  • the PE1 obtains the second data packet according to the first data packet and the first low delay label.
  • the second data packet includes the first data packet.
  • the first service is a low-latency IP service
  • the second data packet further includes the first low-latency tag and the first tag, the first low-latency tag and the first tag Encapsulating is in the outer layer of the first data packet.
  • the first label is a label that is allocated by the PE1 and corresponding to the first LSP. The first label is used to instruct PE1 to forward data packets along the first LSP.
  • the second data packet further includes the first low-latency tag and the first segment routing (segment routing, The SR) tag, the first low latency tag and the first SR tag are encapsulated in an outer layer of the first data packet.
  • the first SR tag is used to identify a link between PE1 and P1.
  • the first label and the first SR label are in the form of two labels, and the first label and the first SR label are used to instruct the PE1 to forward the data packet.
  • the second data packet may further include an ACH header and/or a pseudo wire (PW) tag.
  • PW pseudo wire
  • the PW tag is used to implement Layer 2 Ethernet services.
  • Figure 5 (b) and Figure 5 (c) are used in the second data message A schematic diagram of possible package formats. As shown in FIG. 5(b), the first low latency tag may be packaged between the PW tag and the first tag. Or as shown in FIG. 5(c), the first low latency tag may be encapsulated between the ACH header and the PW tag.
  • the embodiment of the present application does not limit the possible location of the first low-latency label without affecting the forwarding of the second data packet.
  • the PE1 After the PE1 obtains the second data packet, the PE1 sends the second data packet to the to-be-sent queue on the PE1 according to the first low-latency label carried in the second data packet.
  • the to-be-sent queue on PE1 is the priority queue corresponding to the first service flow on PE1.
  • the first low delay label is further configured to indicate that the second data packet belongs to the first service flow.
  • the PE1 sends the second data packet to the P1 in the first low-latency forwarding mode.
  • PE1 can configure the first low-latency forwarding mode by using the method provided in Embodiment 1. This ensures that PE1 has started running the first low-latency forwarding before or at the same time that PE1 sends the two data packets.
  • the mode that is, the gating for forwarding the first service flow on PE1 is in an open state.
  • PE1 can configure a normal gating list through the management plane.
  • PE1 can implement fast processing of low-latency services according to whether the first low-latency forwarding mode is enabled or disabled.
  • the control is flexible and the management is relatively simple.
  • the PE1 sends the second data packet to the P1.
  • the PE1 sends the packet in the to-be-sent queue on the PE1 in the first low-latency forwarding mode. And performing the identification to obtain the second data packet; the PE1 preferentially selecting the second data packet to send according to the first low delay label included in the second data packet.
  • the PE1 uses the credit shaping algorithm, the PE1 preferentially selects the second data packet to be sent, and further includes: PE1 lowering the value of the credit.
  • the to-be-forwarded team on the PE1 may further include a plurality of data packets belonging to the first service flow, and the PE1 may continue to preferentially select the plurality of the first service flows.
  • the data packet is sent until the data packet belonging to the first service flow in the to-be-forwarded team on the PE1 is sent or the value of the credit is reduced to 0.
  • the P1 obtains the third data packet according to the second data packet from the PE1.
  • P1 obtains the third data packet according to the second data packet from the PE1, and the P1 receives the second data packet sent by the PE1, and the P1 obtains the second data packet according to the second data packet.
  • Three data messages If the second data packet includes the first label, P1 replaces the first label included in the second data packet with a second label, and obtains the third data packet.
  • the second label is a label that is allocated by P1 and corresponding to the first LSP. The second label is used to instruct P1 to send the third data packet along the first forwarding path.
  • the first SR tag is located in the first If the data packet includes the top of the stack of the label stack, P1 may pop up the first SR label to obtain the third data packet. That is, the second SR tag is located at the top of the stack of the label stack included in the third data packet.
  • the first SR tag can also identify the link between P1 and PE2, P1 does not process the first SR tag included in the second data packet, and the second data packet is processed.
  • the second label and the second SR label are in two label forms. The second label and the second SR label are both used to indicate that the P1 forwards the data message.
  • P1 After P1 obtains the third data packet, P1 sends the third data packet to the to-be-sent queue in P1 according to the first low-latency label, and the to-be-sent queue on the P1 It is a priority queue corresponding to the first service flow on P1.
  • the P1 sends the third data packet to the PE2 in the first low-latency forwarding mode.
  • P1 may configure the first low-latency forwarding mode by using the method provided in Embodiment 1. This ensures that P1 has started to run the first low-latency forwarding before or at the same time as P1 sends the three data packets.
  • the mode that is, the gating for forwarding the first service flow on P1 is in an open state.
  • P1 can configure a normal gating list through the management plane.
  • P1 can implement fast processing of low-latency services according to whether the first low-latency forwarding mode is turned on or off.
  • the control is flexible and the management is relatively simple.
  • the sending, by the P1, the third data packet to the PE2 in the first low-latency forwarding mode includes: P1, in the first low-latency forwarding mode, the packet in the to-be-sent queue on P1. Identifying, obtaining the third data message; P1 according to the first The first low-latency tag included in the three data packets preferentially selects the third data packet for transmission.
  • the P1 adopts the credit shaping algorithm, the P1 preferentially selects the third data packet to be sent, and further includes: P1 lowering the value of the credit.
  • the to-be-forwarded team on the P1 may further include a plurality of data packets belonging to the first service flow, and P1 may continue to preferentially select the plurality of the first service flows.
  • the data packet is sent until the data packet belonging to the first service flow in the to-be-forwarded team on the P1 is sent or the value of credit is reduced to zero.
  • the PE2 obtains the first data packet according to the third data packet from the P1.
  • the PE2 obtains the first data packet according to the third data packet from the P1, the PE2 receives the third data packet sent by the P1, and the PE2 receives the third data packet from the third data packet. Deleting the first tag and the first low delay tag to obtain the first data packet. After the PE2 obtains the first data packet, the PE2 sends the first data packet to the to-be-sent queue on the PE2 according to the first low-latency label included in the third data packet.
  • the to-be-sent queue on PE2 is a priority queue corresponding to the first service flow on PE2. If the third data packet further includes a PW label and/or an ACH header, the PE2 further deletes the PW label and/or the ACH header during the process of obtaining the first data packet according to the third data packet. .
  • the PE2 sends the first data packet to the CE3 in the first low-latency forwarding mode.
  • PE2 may configure the first low-latency forwarding mode by using the method provided in Embodiment 1. This ensures that PE2 has started running the first low-latency forwarding before or at the same time as PE2 sends the data packet.
  • the mode that is, the gating on the PE2 for forwarding the first service flow is in an open state.
  • PE2 can configure the normal gating list through the management plane.
  • PE2 can implement fast processing of low-latency services according to the first low-latency forwarding mode.
  • the control is flexible and the management is simple.
  • the PE2 sends the first data packet to the CE3, and the PE2 preferentially selects the to-be-transmitted queue on the PE2 in the first low-latency forwarding mode.
  • the first data packet is sent, that is, the first data packet is preferentially sent to the CE3.
  • the PE2 preferentially selects the first data packet to be sent, and further includes: PE2 lowering the value of the credit.
  • the to-be-forwarded team on the PE2 may further include a plurality of data packets belonging to the first service flow, and the PE2 may continue to preferentially select the plurality of the first service flows.
  • the data packet is sent until the data packet belonging to the first service flow in the to-be-forwarded team on the PE2 is sent or the value of the credit is reduced to 0.
  • the method provided in the second embodiment of the present application further includes:
  • PE1 may send the first control information including the first open identifier to P1 by using the method provided in Embodiment 1.
  • first control information including the first open identifier refer to the corresponding content in the first embodiment, and details are not described herein again.
  • PE1 may send the first close identifier to P1 by using the method provided in Embodiment 1.
  • the first shutdown identifier is used to indicate that the forwarding device that receives the first shutdown identifier stops processing the data packet of the first service flow in the first low-latency forwarding mode.
  • the PE1 stops sending the data packet of the first service flow in the first low-latency forwarding mode.
  • the embodiment of the present application may trigger the PE1 to stop sending the data packet of the first service flow in the first low-latency forwarding mode in multiple manners.
  • the PE1 may stop sending the data packet of the first service flow in the first low-latency forwarding mode according to the first shutdown identifier.
  • the PE1 as the ingress node may stop sending the data packet of the first service flow in the first low-latency forwarding mode after determining to complete the sending of the last data packet of the first service flow. .
  • the PE1, which is the ingress node stops sending the data packet of the first service flow in the first low-latency forwarding mode under the control of other control devices or management devices.
  • the PE1 that is the ingress node may stop transmitting the data packet of the first service flow in the first low-latency forwarding mode after receiving the data packet of the first service flow within a predetermined duration. .
  • the P1 stops sending the first service flow in the first low-latency forwarding mode according to the first shutdown identifier sent by the PE1. Data message.
  • the PE2 stops sending the data packet of the first service flow in the first low-latency forwarding mode according to the first shutdown identifier sent by the P1.
  • 409, 410, and 411 may be performed between any of the steps included in the method provided by the second embodiment, and the execution of 409, 410, and 411 depends on the transmission timing of the shutdown identifier.
  • the data message for stopping the sending of the first service flow in the first low-latency forwarding mode may be in the following manner 1 or 2.
  • Manner 1 The first low-latency forwarding mode is disabled, and the data packet of the first service flow obtained after the first low-latency forwarding mode is disabled is sent to a low-priority queue.
  • the low priority queue and the high priority queue are relative concepts, and the low priority queue is a normal queue relative to the priority queue in the first low latency forwarding mode.
  • Manner 2 The first low-latency forwarding mode is disabled, and the data packet of the first service flow obtained after the first low-latency forwarding mode is disabled is discarded.
  • the closing the first low-latency forwarding mode may be performed after obtaining the shutdown identifier.
  • the PE1 may flexibly control the gating associated with the first service flow on P1 and PE2, that is, send the first control for controlling the first low-latency forwarding mode.
  • Information eliminates the need to configure a gated list for each port through the management plane, reducing management complexity.
  • the PE1 sends a data packet of the second service flow to the PE2 along the second LSP, and the PE1 passes the P1 along the first LSP.
  • the method of transmitting the data packet belonging to the first service flow to the PE2 is similar, except that the processing of P1 is not required.
  • the second LSP is a path indicated by a broken line when the carrier network in FIG. 2 is an MPLS network.
  • the second embodiment of the present application is to distinguish the first low-latency label from the label corresponding to the first LSP, that is, each forwarding device on the first LSP is assigned a label for forwarding and the label
  • the first low latency tag is a different tag.
  • the label allocated to each forwarding device on the first LSP for guiding forwarding such as an SR label, a segment label, a first label, or a second label may be extended. Having both the function of indicating how to forward the first service flow (ie, indicating how to forward data messages belonging to the first service flow), and the function of the first low latency tag.
  • the first label and the second label may be extended by the embodiment of the present application, so that the first label and the second label further have the function of the first low delay label.
  • PE1 may not need to add the first low-latency label to the first data packet, and P1 may determine, according to the first label in the second data packet, that the first low-latency forwarding mode is used.
  • the second data packet, the PE2 may determine, according to the second label in the third data packet, that the third data packet is processed by using the first low-latency forwarding mode, thereby further reducing the MPLS network.
  • the complexity of the configuration of the forwarding device may be extended by the embodiment of the present application, so that the first label and the second label further have the function of the first low delay label.
  • PE1 may not need to add the first low-latency label to the first data packet
  • P1 may determine, according to the first label in the second data packet, that the first low-latency forwarding mode is used.
  • the second data packet, the PE2 may determine, according to the
  • FIG. 6 is a flowchart of a method for processing a low-latency service flow according to Embodiment 3 of the present application.
  • the carrier network is the IP network
  • the low-latency service flow is the second service flow
  • the method for processing the low-latency service flow is described.
  • PE1 and PE2 in Figure 2 operate the Internet Protocol (IP).
  • IP Internet Protocol
  • PE1 and PE2 complete the configuration of the second low-latency forwarding mode before processing the second service flow.
  • IP Internet Protocol
  • the second forwarding path between PE1 and PE2 can be pre-planned according to the IP routing protocol, such as Intermediate System to Intermediate System (IS-IS) traffic engineering (TE) or open shortest path priority traffic.
  • IS-IS Intermediate System to Intermediate System
  • TE traffic engineering
  • open shortest path priority traffic open shortest path priority traffic.
  • the planning process of the second forwarding path is not described in the third embodiment of the present application.
  • a method for processing a low-latency service flow provided in Embodiment 3 of the present application will be described below with reference to FIG. 2 and FIG.
  • the PE1 receives the first data packet.
  • PE1 receives the first data packet through the second port.
  • the first data message may be from CE2.
  • the first data The message includes a third MAC address and a third IP address.
  • the third MAC address is the MAC address of CE2.
  • the third IP address is an IP address of CE2.
  • the first data message may further include a fourth MAC address and a fourth IP address.
  • the fourth MAC address is the MAC address of CE4.
  • the fourth IP address is an IP address of CE4.
  • CE2 and CE4 can belong to the same VPN.
  • the first data message may be an IP packet or an Ethernet frame.
  • the method provided in Embodiment 3 of the present application further includes: PE1 determining whether the first data packet belongs to the second service flow; and if the first data packet belongs to the second Traffic flow, PE1 executes 602. If the first data packet does not belong to the second service flow, the PE1 processes the first data packet into the to-be-forwarded queue, and processes the packet in the forwarding queue according to a normal scheduling method.
  • the general scheduling method in the embodiment of the present application is the same as the normal scheduling method in the second embodiment, and details are not described herein again.
  • PE1 may use the method in 401 of the second embodiment to determine whether the first data packet belongs to the second service flow, and may adopt any one of the first mode to the fourth mode.
  • the first mode may include one or more of the second port, the third MAC address, the third IP address, the fourth MAC address, and the fourth IP address.
  • the PE1 may determine, according to the multi-group, whether the first data packet belongs to the second service flow.
  • the feature information table may be stored on the PE1, and the feature information table may include the identifier of the second service flow and the feature information corresponding to the data packet of the second service flow.
  • the feature information includes one or more of a MAC address, an IP address, an application layer port number, VLAN information, VXLAN information, and a physical port.
  • PE1 queries the feature information table with the multi-group. If there is information matching the multi-group in the feature information table, PE1 determines that the first data packet belongs to the second service flow.
  • the matching in the third embodiment of the present application has the same meaning as the matching in the second embodiment, and details are not described herein again.
  • the second mode the feature information table saved on the PE1 includes the identifier of the second service and the feature information corresponding to the data packet belonging to the second service.
  • the feature information corresponding to the data packet belonging to the second service may be the same as the feature information corresponding to the data packet belonging to the second service flow in the first mode.
  • the method for determining, by the PE1, whether the first data packet belongs to the second service flow is the same as the first method according to the feature information corresponding to the data packet that belongs to the second service, and is not specific to the first method. The method of judgment is described in detail.
  • the third mode the first data packet further includes an identifier of the service flow to which the first data packet belongs, and if the identifier of the service flow to which the first data packet belongs is the second service flow And identifying, the PE1 determines that the first data packet belongs to the second service flow.
  • the fourth mode the first data packet further includes an identifier of the service to which the first data packet belongs, and if the identifier of the service to which the first data packet belongs is the identifier of the second service, The PE1 determines that the first data packet belongs to the second service, that is, the first data packet belongs to the second service flow.
  • the PE1 determines that the first data packet belongs to the second service flow.
  • PE1 may obtain a second low-latency identifier after determining that the first data packet belongs to the second service flow.
  • the manner in which the PE1 obtains the second low-latency identifier may be any one of the first mode and the third mode.
  • the PE1 may obtain the identifier of the second service flow after determining that the first data packet belongs to the second service flow according to the feature information table in 601. Or the first data packet carries the identifier of the second service flow, and the PE1 obtains the identifier of the second service flow from the first data packet.
  • a correspondence between the second low-latency identifier and the identifier of the second service flow is configured on the PE1.
  • the PE1 may obtain the second low-latency identifier according to the correspondence and the identifier of the second service flow.
  • the PE1 may obtain the identifier of the second service after determining that the first data packet belongs to the second service flow. Or the first data packet carries the identifier of the second service, and the PE1 obtains the identifier of the second service from the first data packet.
  • a correspondence between the second low-latency identifier and the identifier of the second service is configured on the PE1.
  • the PE1 may obtain the second low-latency identifier according to the correspondence and the identifier of the second service.
  • the PE1 is indeed After the first data packet belongs to the second service flow, the second low delay identifier is directly obtained.
  • the PE1 obtains the second data packet according to the first data packet and the second low-latency identifier.
  • the second data packet includes the first data packet and a tunnel encapsulation.
  • the tunnel encapsulation may be encapsulated in an outer layer of the first data packet.
  • the tunnel encapsulation includes the second low latency identifier.
  • the tunnel encapsulation may be a Generic Routing Encapsulation (GRE) encapsulation.
  • GRE Generic Routing Encapsulation
  • the PE1 After the PE1 obtains the second data packet, the PE1 sends the second data packet to the to-be-sent queue on the PE1 according to the second low-latency identifier, and the to-be-sent queue on the PE1 A priority queue corresponding to the second service flow on PE1.
  • the PE1 sends the second data packet to the PE2 in the second low-latency forwarding mode.
  • PE1 may configure the second low-latency forwarding mode by the method provided in Embodiment 1. In this way, it can be ensured that the PE1 has started to run the second low-latency forwarding mode before the PE1 sends the two data packets, that is, the gating for forwarding the second service flow on the PE1 is in an open state. PE1 can configure the normal gating list through the management plane. The second low-latency forwarding mode can be enabled or disabled to implement fast processing of low-latency services. The control is flexible and the management is simple.
  • the PE1 sends the second data packet to the PE2, and the PE1 identifies the packet in the sending queue in the second low-latency forwarding mode.
  • the second data packet is sent by the PE1 according to the second low-latency identifier included in the second data packet, and the second data packet is preferentially selected for transmission.
  • the PE1 can send the second data packet to the PE2 along the tunnel between the PE1 and the PE2.
  • the PE2 obtains the first data packet according to the second data packet from the PE1.
  • the PE2 obtains the first data packet according to the second data packet from the PE1, and the PE2 receives the second data packet sent by the PE1.
  • the PE2 receives the second data packet from the second data packet. Deleting the tunnel encapsulation to obtain the first data packet.
  • the PE2 sends the first data packet to the to-be-sent queue on the PE2 according to the second low-latency identifier included in the second data packet.
  • the to-be-sent queue on PE2 is a priority queue corresponding to the second service flow on PE2.
  • the PE2 sends the first data packet to the CE4 in the second low-latency forwarding mode.
  • PE2 may configure the second low-latency forwarding mode by the method provided in Embodiment 1. In this way, it can be ensured that the PE2 has started to run the second low-latency forwarding mode before the PE2 sends the data packet, that is, the gating for forwarding the second service flow on the PE2 is in an open state. PE2 can configure the normal gating list through the management plane. The second low-latency forwarding mode can be enabled or disabled to implement fast processing of low-latency services. The control is flexible and the management is simple.
  • the PE2 sends the first data packet to the CE4, and the PE2 preferentially selects the to-be-transmitted queue in the PE2 in the second low-latency forwarding mode.
  • the first data packet is sent, that is, the first data packet is preferentially sent to the CE4.
  • the method provided in the third embodiment of the present application further includes :
  • PE1 may send the second control information including the second open identifier to P1 by using the method provided in Embodiment 1.
  • the second control information including the second open identifier refer to the corresponding content in the first embodiment, and details are not described herein again.
  • PE1 may send a second close identifier to PE2 by using the method provided in Embodiment 1.
  • the second closing identifier is used to indicate that the forwarding device that receives the second closed identifier stops processing the data packet of the second service flow in the second low-latency forwarding mode.
  • the PE1 stops sending the data packet of the second service flow in the second low-latency forwarding mode.
  • the embodiment of the present application may trigger the PE1 to stop sending the data packet of the second service flow in the first low-latency forwarding mode in multiple manners.
  • the PE1 may stop sending the the low-latency forwarding mode in the second low-delay forwarding mode according to the second shutdown identifier.
  • the data packet of the second service flow In the second mode, the PE1 as the ingress node may stop sending the data packet of the second service flow in the second low-latency forwarding mode after determining to complete the sending of the last data packet of the second service flow. .
  • the PE1, which is the ingress node stops sending the data packet of the second service flow in the second low-latency forwarding mode under the control of other control devices or management devices.
  • the PE1, which is the ingress node stops transmitting the data packet of the second service flow in the second low-latency forwarding mode after receiving the data packet of the second service flow within a predetermined duration. .
  • the PE2 stops sending the data packet of the second service flow in the second low-latency forwarding mode according to the received second closed identifier from the PE1.
  • the data packet for stopping the sending of the second service flow in the second low-latency forwarding mode may adopt mode one or mode two.
  • Manner 1 The second low-latency forwarding mode is turned off, and the data packet of the second service flow obtained after the second low-latency forwarding mode is disabled is sent to the low-priority queue.
  • the low priority queue and the high priority queue are relative concepts, and the low priority queue is a normal queue relative to the priority queue in the second low latency forwarding mode.
  • Manner 2 The second low-latency forwarding mode is disabled, and the data packet of the second service flow obtained after the second low-latency forwarding mode is disabled is discarded.
  • the closing the second low-latency forwarding mode may be after obtaining the shutdown identifier.
  • the PE1 may flexibly control the gating associated with the second service flow on the PE2, that is, send the second control information for controlling the second low-latency forwarding mode. Reduce the complexity of management by eliminating the need to configure a gated list for each port through the management plane.
  • the PE1 sends a data packet belonging to the first service flow to the PE2 through the P1 along the first forwarding path, and the PE1 follows the second forwarding.
  • the method of sending the data packet belonging to the second service flow to the PE2 is similar to the path, and is not described here.
  • the first forwarding path is a path indicated by a solid line when the carrier network in FIG. 2 is an IP network.
  • FIG. 7 is a flowchart of a method for detecting a transmission delay of a forwarding path according to Embodiment 4 of the present application.
  • the fourth embodiment of the present application is a method for detecting a low-latency service flow of a forwarding path by taking a first forwarding path as an example.
  • the method provided in Embodiment 4 may be performed after the device on the first forwarding path sets the first low-latency forwarding mode, and before the device on the first forwarding path processes the first service flow.
  • a method for detecting a transmission delay of a forwarding path according to Embodiment 4 of the present application will be described below with reference to FIG. 2 and FIG.
  • the PE1 sends a first detection packet to the P1.
  • the first detection packet includes a first delay value and a second delay value.
  • the first detection packet is used to obtain a transmission delay generated by the forwarding device on the first forwarding path.
  • the first delay value is used to indicate a maximum duration of delay allowed by the forwarding device on the first forwarding path.
  • the second delay value is a delay time generated by PE1.
  • the delay time generated by PE1 may be obtained by actual measurement calculation to obtain a delay value generated by PE1 forwarding the first detection message.
  • the first detection packet further includes a time when the PE1 sends the first detection packet.
  • the forwarding device that is the egress of the first forwarding path such as the PE2, obtains the first forwarding path according to the time when the first detection packet is sent by the PE1 and the time when the second detection packet is received by the PE2. Transmission delay.
  • the method for the PE1 to send the first detection packet is the same as the method for the PE1 to send the second data packet in the second embodiment or the third embodiment, and details are not described herein.
  • the P1 obtains the second detection packet according to the first detection packet.
  • the P1 obtains the second detection packet according to the first detection packet, where the P1 obtains a third delay value, and the third delay value is P1, and the first detection packet is sent to the sending station. Describe the duration between the first detection messages; P1 is based on the first detection message and the And the third detection packet is obtained by the third detection packet, where the second detection packet includes the first detection packet and the third delay value.
  • the time at which the first detection packet is sent by the P1 is the time at which the second detection packet is sent by the P1, and the time at which the first detection packet is sent is obtained by P1.
  • the P1 sends the second detection packet to the PE2.
  • the method for sending the second detection packet by the P1 may be the same as the method for sending the third data packet by the P1 in the second embodiment or the third embodiment, and details are not described herein again.
  • the PE2 obtains a third detection packet according to the second detection packet.
  • the PE2 obtains the third detection packet according to the second detection packet, where the PE2 obtains the fourth delay value, and the fourth delay value is the PE2 that receives the second detection packet to the sending station. Determining the duration of the second detection packet; the PE2 obtaining the third detection packet according to the second detection packet and the fourth delay value, where the third detection packet includes the second detection The message and the fourth delay value.
  • the time at which the second detection packet is sent by the PE2 is the time at which the PE2 sends the third detection packet, and the PE2 can estimate the time at which the second detection packet is sent.
  • the method provided in Embodiment 4 of the present application further includes: PE2 reporting the third detection packet.
  • the PE2 may send the third detection packet to the device for performing management, or the PE2 may send the information and/or parameters carried in the third detection packet to the device for performing management.
  • the device for performing management obtains a transmission delay of the first forwarding path.
  • the device for management may be a network management platform.
  • the PE1 sends the first detection packet along the first forwarding path, which facilitates processing the first detection packet by the device on the first forwarding path to obtain each device.
  • the resulting delay value is used to locate the node that cannot satisfy the first delay value, which helps to obtain the best forwarding effect of low-latency services.
  • FIG. 8 is a schematic diagram of a first forwarding device according to Embodiment 5 of the present application.
  • the first forwarding device provided in Embodiment 5 of the present application may be PE1 in FIG. 3, FIG. 4 or FIG.
  • the first forwarding device provided in Embodiment 5 of the present application may adopt a method adopted by PE1.
  • the first forwarding device provided in Embodiment 5 of the present application is described below with reference to FIG.
  • the first forwarding device provided in Embodiment 5 of the present application includes a processing unit 801, an obtaining unit 802, and a first sending unit 803.
  • the processing unit 801 is configured to determine that the received first data packet belongs to the first service flow, and obtain a low delay identifier corresponding to the first service flow.
  • the obtaining unit 802 is configured to obtain, after the processing unit 801 determines that the first data packet belongs to the first service flow, obtain the second datagram according to the first data packet and the low delay identifier. Text.
  • the second data packet includes the first data packet and the low latency identifier.
  • the low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow under dynamic control.
  • the low-latency identifier is used to indicate that the forwarding device that receives the first service flow forwards the first service flow in a low-latency forwarding mode.
  • the second data packet belongs to the first service flow.
  • the first sending unit 803 is configured to send the second data packet to the second forwarding device in the low-latency forwarding mode. If the first forwarding device sends the second data along the first forwarding path, the second forwarding device in this embodiment may be P1 in FIG. 3 or FIG. 4, where P1 is the first forwarding path of the PE1. The next hop on. If the first forwarding device sends the second data along the second forwarding path, the second forwarding device in this embodiment may also be the PE2 in FIG. 3 or FIG. 6, and the PE2 is the second forwarding in the PE1. The next hop on the path.
  • the obtaining unit 802 is further configured to obtain control information.
  • the first forwarding device further includes: a second sending unit 804.
  • the second sending unit 804 is configured to send the control information to the second forwarding device, where the control information is used to control a state of the low-latency forwarding mode.
  • the content of the control information can be referred to the corresponding content in the first embodiment.
  • control information includes an open identifier, where the open identifier is used to identify that the low-latency forwarding mode is enabled, and the first forwarding device further includes a third sending unit 805.
  • the third sending unit 805 is configured to send a shutdown identifier to the second forwarding device, where the shutdown identifier is used to identify that the low-latency forwarding mode is disabled.
  • the obtaining unit 802 is further configured to obtain a detection message.
  • the first forwarding device further includes: a fourth sending unit 806.
  • the fourth sending unit 806 is configured to send the detection packet to the second forwarding device, where the detection packet includes a first delay value and a second delay value, where the detection packet is used to obtain a forwarding path.
  • the transmission delay of the forwarding device, the first delay value is a maximum duration of delay allowed by the forwarding device on the forwarding path, and the second delay value is a delay duration generated by the first forwarding device.
  • the first forwarding device may respectively follow the first The forwarding path sends a detection packet, and another detection packet is sent along the second forwarding path.
  • the first forwarding device provided by the embodiment of the present application can flexibly control the gating associated with the first service flow on the second forwarding device, that is, send the first control information for controlling the low-latency forwarding mode, without passing through
  • the management plane configures a gated list for each port, reducing management complexity.
  • FIG. 9 is a schematic diagram of a second forwarding device according to Embodiment 6 of the present application.
  • the second forwarding device provided in Embodiment 6 of the present application may be P1 of FIG. 3 or FIG. 4, where P1 is the next hop of PE1 on the first forwarding path.
  • the second forwarding device may also be the PE2 of FIG. 3 or FIG. 4, and the PE2 is the next hop of the PE1 on the second forwarding path.
  • the second forwarding device provided in Embodiment 6 of the present application may adopt the method adopted by P1 or PE2, and the same content as in any one of Embodiments 1 to 4 will not be described herein.
  • the second forwarding device provided in Embodiment 6 of the present application is described below with reference to FIG.
  • the second forwarding device includes: a first receiving unit 901 and a first sending unit 902.
  • the first receiving unit 901 is configured to receive a second data packet from the first forwarding device.
  • the second data packet includes a first data packet and a low latency identifier.
  • the low-latency forwarding mode is a mode for implementing fast forwarding of the first service flow under dynamic control.
  • the low-latency identifier is used to indicate that the forwarding device that receives the first service flow forwards the first service flow in a low-latency forwarding mode.
  • the second data packet belongs to the first service flow.
  • the first sending unit 902 is configured to send the second data packet in a low delay forwarding mode according to the low delay identifier. If the second forwarding device is an intermediate forwarding device, such as P1, the first sending unit 902 may send the second data packet to the next hop in the low-latency forwarding mode. If the second forwarding device is a node that is a network egress, such as PE2, the first sending unit 902 may send the second data packet to the CE that communicates with the CE in the low-latency forwarding mode. The first data message.
  • the second forwarding device further includes: a second receiving unit 903 and a control unit 904.
  • the second receiving unit 903 is configured to receive control information from the first forwarding device.
  • the control information is used to control the state of the low latency forwarding mode.
  • the control unit 904 is configured to control a state of the low-latency forwarding mode according to the control information.
  • control information includes a start time and an end time of the low-latency forwarding mode.
  • the control unit 904 is specifically configured to run the low-latency forwarding mode according to a start time and an end time of the low-latency forwarding mode.
  • control information includes a start time and a running time of the low-latency forwarding mode.
  • the control unit 904 is specifically configured to run the low-latency forwarding mode according to a start time of the low-latency forwarding mode and the running duration.
  • control information includes an opening identifier.
  • the opening identifier is used to identify that the low-latency forwarding mode is enabled.
  • the second forwarding device further includes: a third receiving unit 905.
  • the third receiving unit 905 is configured to receive the shutdown identifier sent by the first forwarding device.
  • the shutdown identifier is used to identify that the low latency forwarding mode is turned off.
  • the control unit 904 is further configured to control according to the closing identifier
  • the first sending unit 902 stops transmitting the data packet of the one service flow in the low delay forwarding mode.
  • the second forwarding device further includes: a fourth receiving unit 906, a first obtaining unit 907, and a second obtaining unit 908.
  • the fourth receiving unit 906 is configured to receive a first detection packet from the first forwarding device, where the first detection packet includes a first delay value and a second delay value, and the first detection packet The method is used to obtain a transmission delay generated by the forwarding device on the forwarding path, where the first delay value is a maximum duration of delay allowed by the forwarding device on the forwarding path, and the second delay value is the first The delay time generated by a forwarding device.
  • the first obtaining unit 907 is configured to obtain a third delay value, where the third delay value is that the second forwarding device receives the first detection packet to send the first detection packet. duration.
  • the second obtaining unit 908 is configured to obtain a second detection packet according to the first detection packet and the third delay value, where the second detection packet includes the first detection packet and the The third delay value is described.
  • the second forwarding device is an intermediate forwarding device, where the first sending unit 902 is specifically configured to send the second detection packet to the third forwarding device, where the third forwarding device is the forwarding path.
  • the second forwarding device in the second embodiment of the present application may send the second data packet in a low-latency forwarding mode according to the configured low-latency forwarding mode after receiving the first data packet, without storing the management plane.
  • the issued gating list provides more flexible control and helps to reduce the complexity of management.
  • FIG. 10 is a schematic diagram of a first forwarding device according to Embodiment 7 of the present application.
  • the first forwarding device shown in FIG. 10 and the first forwarding device shown in FIG. 8 may be the same device.
  • the first forwarding device shown in FIG. 10 may be the PE1 in any one of the foregoing Embodiments 1 to 4.
  • the first forwarding device provided in Embodiment 7 of the present application includes: a processor 1001, a memory 1002, and a communication interface 1003.
  • the processor 1001, the memory 1002, and the communication interface 1003 are connected by a communication bus 1004.
  • the memory 1002 is used to store programs. Alternatively, the memory 1002 may also be used to store control information and/or a feature information table in the second embodiment.
  • the processor 1001 performs the following operations according to executable instructions included in the program read from the memory 1002:
  • the forwarding device that receives the first service flow forwards the first service flow in a low-latency forwarding mode, where the second data packet belongs to the first service flow;
  • the processor 1001 may further send control information to the second forwarding device by using the communication interface 1003 according to the executable instruction.
  • the control information is used to control the state of the low latency forwarding mode.
  • control information is carried in an RSVP message or a G-ACH channel message.
  • RSVP message and G-ACH channel message refer to the corresponding content in the first embodiment, and details are not described herein again.
  • control information can be referred to the corresponding content in the first embodiment, and details are not described herein again.
  • the processor 1001 may further send, according to the executable instruction, to the second forwarding device by using the communication interface 1003 before sending the second data packet to the second forwarding device.
  • the detection message includes a first delay value and a second delay value. The content of the first delay value and the second delay value are the same as those in the foregoing embodiment, and details are not described herein again.
  • the communication interface 1003 includes at least two logical interfaces, at least one queue, and at least one gate.
  • FIG. 10 includes only the first logical interface 10030, the queue 10031, the gate 10032, and the second with the communication interface 1003.
  • the logical interface 10033 is described as an example.
  • the queue 10031, the gate 10032, and the second logical interface 10033 are configured to process data packets of the first service flow.
  • the processor 1001 can receive the first data packet through the first logical interface 10030.
  • the first logical interface 10030 in this embodiment may be the first port or the second port in the second embodiment.
  • the processor 1001 can output the obtained second data message to the Queue 10031.
  • the processor 1001 controls the opening of the gate 10032 according to the control information, that is, switches to the low-latency forwarding mode.
  • the processor 1001 controls a data packet of the first service flow in the queue 10031, for example, the second data packet, and outputs the data packet to the gate 10032.
  • the gate 10032 is in an open state, and the second data packet is sent to the second logical interface 10033 and output to the corresponding physical line through the second logical interface 10033.
  • the communication interface 1003 further includes a queue 10032 capable of communicating with the second logical interface 10033.
  • the queue 10032 is a queue formed by the data packet to be forwarded processed in the normal forwarding mode. After the second forwarding device stops using the low-latency forwarding mode, the queue 10032 and the second logical interface 10033 can use the normal queue forwarding manner to send data packets.
  • the priority of the queue 10032 is lower than the priority of the queue 10031.
  • the second logical interface 10033 and the first logical interface 10030 may be physically the same physical interface, and the physical interface can implement a transceiving function.
  • the second logical interface 10033 and the first logical interface 10030 are physically implemented by different physical interfaces, and the specific implementation manner of the embodiment of the present application is not limited.
  • the processor 1001 can control the opening of the gate 10032 according to its internal clock and the control information.
  • the first forwarding device further includes a timer 1005.
  • a clock 1005 can be used to synchronize the clocks of the processor 1001 and the gate 10032.
  • the processor 1001 can control the opening of the gate 10032 according to the clock 1005 and the control information.
  • FIG. 11 is a schematic diagram of a second forwarding device according to Embodiment 8 of the present application.
  • the second forwarding device shown in FIG. 11 and the second forwarding device shown in FIG. 9 may be the same device.
  • the second forwarding device shown in FIG. 11 may be P1 or PE2 in any one of Embodiments 1 to 4 above.
  • the second forwarding device provided in Embodiment 8 of the present application includes: a processor 1101, a memory 1102, and a communication interface 1103.
  • the processor 1101, the memory 1102, and the communication interface 1103 are connected by a communication bus 1104.
  • the memory 1102 is used to store programs. Alternatively, the memory 1102 can also be used to store control information.
  • the processor 1101 performs the following operations according to executable instructions included in the program read from the memory 1102:
  • the forwarding device of the service flow forwards the first service flow in a low-latency forwarding mode, where the second data packet belongs to the first service flow;
  • the processor 1101 may further receive, by using the communication interface 1103, control information from the first forwarding device according to the executable instruction.
  • the control information is used to control the state of the low latency forwarding mode.
  • the processor 1101 further controls a state of the low-latency forwarding mode according to the control information.
  • control information may be carried in an RSVP message or a G-ACH channel message.
  • RSVP RSVP message
  • G-ACH channel message For details, refer to the corresponding content in Embodiment 1.
  • the content of the control information can be referred to the corresponding content in the first embodiment, and details are not described herein again.
  • the method for the processor 1101 to dynamically control the low-latency forwarding mode according to the control information refer to the corresponding content in the first embodiment or the second embodiment, and details are not described herein again.
  • the processor 1101 may further receive, by using the communication interface 1103, the first detection packet from the first forwarding device according to the executable instruction.
  • the first detection packet includes a first delay value and a second delay value.
  • the processor 1101 further obtains a third delay value.
  • the first delay value, the second delay value, and the third delay value are the same as those in the foregoing embodiment, and details are not described herein again.
  • the processor 1101 further obtains a second detection packet according to the first detection packet and the third delay value.
  • the second detection packet includes the first detection packet and the third delay value.
  • the second forwarding device is an intermediate forwarding device, such as P1 in Embodiment 1, Embodiment 2, or Embodiment 4,
  • the processor 1101 may further send the second detection message to the third forwarding device by using the communication interface 1103 according to the executable instruction.
  • the third forwarding device is a next hop of the second forwarding device on the forwarding path.
  • the communication interface 1103 includes at least two logical interfaces, at least one queue, and at least one gate.
  • FIG. 11 includes only the first logical interface 11030, the queue 11031, the gate 11032, and the second with the communication interface 1103.
  • the logical interface 10033 is described as an example.
  • the queue 11031, the gate 11032, and the second logical interface 10033 are configured to process data packets of the first service flow.
  • the processor 1101 can receive the second data packet by using the first logical interface 11030.
  • the processor 1101 may output the second data message to the queue 11031.
  • the processor 1101 controls the opening of the gate 11032 according to the control information, that is, switches to the low-latency forwarding mode.
  • the processor 1101 controls a data message of the first service flow in the queue 11031, for example, the second data message, and outputs the data to the gate 11032 preferentially.
  • the gate 11032 is in an open state, and the second data packet is sent to the second logical interface 11033, and is output to the corresponding physical line through the second logical interface 11033.
  • the communication interface 1103 further includes a queue 11032 capable of communicating with the second logical interface 11033.
  • the queue 11032 is a queue formed by data packets to be forwarded processed in a normal forwarding mode. After the second forwarding device stops using the low-latency forwarding mode, the queue 11032 and the second logical interface 11033 can use the normal queue forwarding manner to send data packets.
  • the priority of the queue 11032 is lower than the priority of the queue 11031.
  • the second logical interface 11033 and the first logical interface 11030 may be physically the same physical interface, and the physical interface can implement the transceiver function.
  • the second logical interface 11033 and the first logical interface 11030 are physically implemented by different physical interfaces, and the specific implementation manner of the embodiment of the present application is not limited.
  • the processor 1101 can control the opening of the gate 11032 according to its internal clock and the control information.
  • the first forwarding device further includes a timer 1105.
  • the processor 1101 can control the opening of the gate 11032 according to the clock 1105 and the control information.
  • a clock 1105 can be used to synchronize the clocks of the processor 1101 and the gate 11032.
  • Embodiment 9 of the present application also provides a system for processing a low-latency service flow.
  • the system provided in Embodiment 9 of the present application may include the first forwarding device provided in Embodiment 5 and the second forwarding device provided in Embodiment 6.
  • the system provided by the embodiment of the present application may include the first forwarding device provided in Embodiment 7 and the second forwarding device provided in Embodiment 8.
  • the function and structure of each forwarding device are not described here. For details, refer to the description in the corresponding embodiment.
  • FIG. 12 is a flowchart of a method for establishing a forwarding path according to Embodiment 10 of the present application.
  • the method provided in Embodiment 10 can be used to establish a first forwarding path or a second forwarding path in the network scenario shown in FIG. 2.
  • the method provided in Embodiment 10 can be performed by any one of the operator networks shown in FIG. 2.
  • the first node in the method provided in Embodiment 10 is any one of the operator networks shown in FIG. 2.
  • the first node may be PE1, P1 or PE2 in FIG. 2.
  • the method provided in Embodiment 10 of the present application can determine the first forwarding path and the second forwarding path shown in FIG. 2 from the network scenario shown in FIG.
  • the method provided in Embodiment 10 of the present application will be described below with reference to the network scenario shown in FIG. 13 and FIG.
  • the first node obtains topology information in a carrier network.
  • the topology information is topology information of a node on each path between the ingress node and the egress node.
  • the topology information includes a physical link delay between two adjacent nodes and a node resident duration of each node.
  • the ingress node in the embodiment of the present application is a node that is a network entry.
  • the egress node in the embodiment of the present application is a node that is a network egress.
  • the first node obtains a transmission delay of each path between the ingress node and the egress node according to the topology information.
  • the first node determines, according to a transmission delay of each path, a target transmission path between the ingress node and the egress node.
  • the ingress node is PE1 and the egress node is PE2.
  • the target transmission path between PE1 and PE2 may be the first forwarding path and/or the second forwarding path mentioned in the above embodiments.
  • the transmission delay of the first forwarding path is the sum of the physical link delay between two adjacent nodes on the first forwarding path and the node resident duration of each node on the first forwarding path.
  • the transmission delay of the first forwarding path is a physical link delay between PE1 and P1, a physical link delay between P1 and PE2, a node resident duration of PE1, and P1.
  • the transmission delay of the second forwarding path is the sum of the physical link delay between two adjacent nodes on the second forwarding path and the node resident duration of each node on the second forwarding path. As shown in FIG. 2, the transmission delay of the second forwarding path is the sum of the physical link delay between PE1 and PE2, the duration of the node resident of PE1, and the duration of the node resident of PE2.
  • the any one node can measure the physical link delay between the node and the neighbor node, and can also measure the node resident of any one of the nodes. duration.
  • the physical link delay obtained by any one of the nodes and the node of any one of the nodes may be obtained in the form of topology information after measuring the physical link delay and the duration of its node resident.
  • the resident duration is sent to other nodes in the carrier network.
  • the node resident duration mentioned in the embodiment of the present application is the delay value generated by the node, or the delay duration generated by the node.
  • the first node may receive topology information sent by P1.
  • the topology information sent by the P1 includes a physical link delay between P1 and PE1, a physical link delay between P1 and PE2, and a node resident duration of P1.
  • the first node may also receive topology information sent by the PE2.
  • the topology information sent by PE2 includes the physical link delay between P1 and PE2, the physical link delay of PE2 and PE1, and the duration of the node of PE2.
  • the first node may receive topology information sent by the PE1.
  • the topology information sent by the PE1 includes a physical link delay between P1 and PE1, a physical link delay between PE1 and PE2, and a node resident duration of PE1.
  • the first node may be a node other than PE1, P1, and PE2 in the carrier network shown in FIG. 2, or may be any one of PE1, P1, and PE2.
  • the first node may determine, according to the obtained topology information, a target transmission path between the ingress node and the egress node.
  • the target transmission path is a path for transmitting a low-latency service flow that meets a delay requirement, such as the first forwarding path or the second forwarding path shown in FIG. 2.
  • the first node obtains the first topology information as an example for description.
  • the first node may perform measurement of the first physical link delay by receiving a delay measurement message from the first neighbor node. For example, the first node receives a delay measurement packet sent by the first neighbor node.
  • the delay measurement packet includes a sending timestamp of the first neighbor node sending the delay measurement packet.
  • the first node may obtain the first physical link delay according to the receiving timestamp of receiving the delay measurement message and the sending timestamp.
  • the delay measurement message is directly sent to the first node by the first neighbor node without being forwarded by other nodes. As shown in FIG. 13, if the first node is the node 4 in FIG. 13 and the first neighbor node is the node 6 in FIG.
  • the node 6 sends a delay measurement message to the node 4, when The delay measurement packet includes a sending timestamp t1 at which the node 6 sends the delay measurement message.
  • the node 4 obtains a reception time stamp t2.
  • the node 4 can obtain the physical link delay between the node 6 and the node 4 according to the receiving timestamp t2 and the sending timestamp t1 included in the delay measurement message.
  • the method for obtaining the physical link delay between any two adjacent nodes in FIG. 13 is the same as the method for obtaining the physical link delay between the node 6 and the node 4, and is not illustrated here.
  • Node 1 in Fig. 13 may be P1 in Fig. 2.
  • the value of the physical link delay between any two adjacent nodes does not vary from direction to direction. If the value of the physical link delay between any two adjacent nodes may be different due to different directions, the physical link delay in different directions between any two adjacent nodes may be detected separately, and The average of the detection results in both directions is taken as the physical link delay between any two adjacent nodes.
  • multiple measurements may be performed by transmitting multiple delay measurement messages. Performing statistics on the multiple measurement results, such as averaging, calculating an expected value, taking a maximum value, or taking a minimum value, etc., to obtain a physical link delay between the two adjacent nodes.
  • the duration of the node staying in the node is closely related to the actual load of the node.
  • the node duration of the message is very small.
  • the node of the message resides for a long time. Therefore, in order to more accurately reflect the node resident duration of the message, in 1201, the node resident duration can be obtained by querying the mapping table between the node load and the resident duration.
  • the length of the node dwell can be determined by the network topology, link medium, length, and the device of the node.
  • a mapping table between the node load and the resident duration of the node may be pre-established in each node. For example, for any one node, the mapping table of the load and the resident duration in the node may be Table 1.
  • Node load percentage
  • Node dwell time 0% 0ms 10% 0.01ms 20% 0.05ms ... ... 50% 0.5ms
  • the node resident duration of the node is 0.05 ms. Any node can obtain its load status, query the mapping table according to its load condition, and obtain the node resident time.
  • the load of any one of the nodes may be a load at any time, or an average value of the load in a certain period of time, or a load at a certain time.
  • any one of the nodes uses the Open Shortest Path First (OSPF) or IS-IS protocol to transmit the topology information obtained by the any one of the nodes.
  • the topology information obtained by any one of the nodes may be topology information obtained by itself, and may also be topology information of other nodes.
  • the OSPF or IS-IS protocol can be extended to carry the topology information obtained by any one of the nodes. The possible structures are not described here.
  • the first node may determine, according to a transmission delay of each path between the ingress node and the egress, that the transmission path that meets the delay requirement is the target transmission path.
  • the target transmission path may be one path, or may be multiple paths that meet the delay requirement.
  • the delay requirement may be that the transmission delay required for the low-latency service flow is less than or equal to a preset value.
  • PE1 is the ingress node
  • PE2 is the egress node
  • the first node may calculate a transmission delay of each transmission path in FIG.
  • the first forwarding path (solid line) in FIG. 2 is a transmission path in which the delay in the plurality of transmission paths in FIG.
  • the second forwarding path (dashed line) in FIG. 2 is a transmission path in which the delay in the plurality of transmission paths in FIG. 13 is small and the delay requirement of the second service flow is met.
  • the first node may determine that the first forwarding path (solid line) and the second forwarding path (dashed line) in FIG. 2 are target transmission paths.
  • the node in the tenth embodiment of the present application may be a device in the carrier network shown in FIG. 2.
  • the above general purpose processor may be a microprocessor or the processor or any conventional processor.
  • the steps of the method disclosed in the embodiment of the present application may be directly implemented as a combination of hardware and software modules in the processor.
  • the code implementing the above functions may be stored in a computer readable medium.
  • Computer readable media includes computer storage media.
  • a storage medium may be any available media that can be accessed by a computer.
  • the computer readable medium may be a random access memory (English full name is random-access memory, abbreviated as RAM in English), read-only memory (English full name is read-only memory, English abbreviation for ROM) , electrically erasable programmable read-only memory (English full name electrically programmable programmable read-only memory, English abbreviation for EEPROM), read-only optical disk (English full name compact disk read-only memory, English abbreviation for CD-ROM) or other Optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.
  • RAM random access memory
  • read-only memory English full name is read-only memory, English abbreviation for ROM
  • electrically erasable programmable read-only memory English full name electrically programmable programmable read-only memory, English abbreviation for EEPROM
  • read-only optical disk English full name compact
  • the computer readable medium may be a compact disc (English full name compact disk, abbreviated as CD), a laser disc, a digital video disc (English full name digital video disc, abbreviated as DVD), a floppy disk or a Blu-ray disc.

Landscapes

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

Abstract

本申请实施例公开了一种用于处理低延迟业务流的方法和装置，有助于降低管理的复杂性和提高控制的灵活性。该方法包括：第一转发设备确定接收的第一数据报文属于第一业务流后，获得与所述第一业务流对应的低延迟标识，所述第一转发设备是作为网络入口的设备；所述第一转发设备根据所述第一数据报文和所述低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式；所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。

Description

一种用于处理低延迟业务流的方法和装置
本申请要求于2016年8月27日提交中国专利局、申请号为CN 201610743336.6、发明名称为“一种用于处理低延迟业务流的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，尤其涉及一种用于处理低延迟业务流的方法和装置。
背景技术
随着高清视频点播的流行以及车载-车载网络(vehicle to vehicle，V2V)的出现，越来越多的业务要求网络传输的时延较低。低延迟业务流随着上述业务的发展和出现而产生。低延迟业务流是一种要求传输时延在预设的阈值内的业务流，其预设的阈值通常较低，比如毫秒级。
目前，以太网络中的设备可以通过管理平面静态配置门控列表(gate control list)来满足低延迟业务流对时延的需求。例如：以太网络的设备的每个以太网端口设置有8个队列，每个队列中按照接收报文的顺序存储有待发送的报文。每个队列对应一个用于控制报文发送的门，即每个以太网端口还设置有8个门。每个以太网端口被配置了一个门控列表。所述门控列表中包括很多个表项，每个表项包括8个门控值。以太网端口包括的8个门可根据每个表项包括的8个门控值开启或关闭。在门被开启的情况下，开启的门对应的队列中的待发送的报文被执行发送动作。采用门控列表进行低延迟业务流的处理，需要对每个以太网端口配置门控列表，管理复杂且不够灵活。
发明内容
有鉴于此，本申请实施例提供一种用于处理低延迟业务流的方法和装置，有助于降低管理的复杂性和提高控制的灵活性。
本申请实施例提供的技术方案如下。
第一方面，提供了一种用于处理低延迟业务流的方法，包括：
第一转发设备确定接收的第一数据报文属于第一业务流后，获得与所述第一业务流对应的低延迟标识，所述第一转发设备为作为网络入口的设备；
所述第一转发设备根据所述第一数据报文和所述低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式；
所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。
其中，所述低延迟业务流为需要传输时延在预设的时长内的业务流。所述预设的时长通常较低，比如毫秒级，可以是10毫秒级别或者5毫秒级别。所述传输时延为网络中端到端所需求的传输时延。所述低延迟业务流可包括一个或多个IP分组，或者所述低延迟业务流可包括一个或多个以太帧。
其中，所述低延迟转发模式是通过指令或信息的交互来实现配置的模式，即动态控制是通过指令或信息的交互来对所述低延迟转发模式的状态进行控制。所述低延迟转发模式为有助于保证低延迟业务流能够得到优先处理和快速转发的模式，比如采用时隙门控功能将门对应的队列中的低延迟业务流进行优先处理。所述时隙门控功能是在预设的时间段内控制所述门开启或关闭的功能。
可选地，所述低延迟标识还用于标识属于所述第一业务流的数据报文，比如所述低延迟标识可用于标识所述第二数据报文属于所述第一业务流。
本申请实施例提供的方法中，第一转发设备在接收到的属于第一业务流的数据报文中插入与所述第一业务流对应的低延迟标识，比如在所述第一业务流的第一数据报文中插入与所述第一业务流对应的低延迟标识，获得第二数据报文。所述第一转发设备在低延迟转发模式下，转发所述第二数据报文，有助于加速属于低延迟业务流的数据报文的转发。所述第一转发设备上可根据所述低延迟转发模式的状态，对所述第一业务流进行转发，无需在所述第一转发设备的每个端口配置门控列表，有助于降低管理的复杂性和提高控制的灵活性。
其中，所述第一转发设备可对所述第一业务流包括的一个或多个数据报文，添加所述低延迟标识，即所述第一业务流包括的一个或多个数据报文经所述第一转发设备处理后，可包括相同的低延迟标识。这样，简化了用于转发所述第一业务流的转发路径上的设备配置，有助于快速识别和处理属于同一业务流的数据报文。所述第一数据报文可以为所述第一业务流的第一个数据报文，或者为所述第一业务流的最后一个数据报文，或者为所述第一业务流中除所述第一个数据报文和所述最后一个数据报文之外的任意一个数据报文。
可选地，在所述第一转发设备确定接收的第一数据报文属于第一业务流之前，所述方法还包括：所述第一转发设备接收所述第一数据报文；所述第一转发设备判断所述第一数据报文是否属于所述第一业务流；所述第一转发设备根据所述判断结果确定所述第一数据报文属于所述第一业务流。
其中，所述第一转发设备判断所述第一数据报文是否属于所述第一业务流包括：所述第一转发设备可根据接收所述第一数据报文的端口或所述第一数据报文携带的信息，判断所述第一数据报文是否属于所述第一业务流。比如所述第一转发设备上可保存有第一业务流的特征信息表，所述特征信息表中可包括用于确定数据报文属于所述第一业务流的特征信息。所述特征信息包括媒体介入控制(Media Access Control，MAC)地址、互联网协议(Internet Protocol，IP)地址、应用层端口号、虚拟局域网(virtual local area network，VLAN)信息、虚拟扩展局域网(Virtual Extensible LAN，VXLAN)信息和物理层端口号中的一个或多个信息。所述第一转发设备可用所述第一数据报文携带的信息或接收所述第一数据报文的端口查询所述特征信息表，若所述特征信息表中存在与所述第一数据报文携带的信息相同的特征信息，或者若所述特征信息表中存在与所述接收所述第一数据报文的端口匹配的物理端口，则所述第一转发设备确定所述第一数据报文属于所述第一业务流。所述第一业务流属于低延迟业务流。
其中，所述第一转发设备接收的所述第一数据报文可以是经用户边缘设备(customer edge，CE)转发的来自所述用户的所述第一数据报文。所述第一转发设备为作为网络入口的设备，比如所述第一转发设备可以是入口(ingress)运营商边缘(provider edge，PE)设备。所述第二转发设备可以是转发路径上的设备，比如所述第二转发设备可以是中间(transit)运营商(provider，P)设备。或者所述第二转发设备可以是作为网络出口的设备，比如所述第二转发设备可以是出口(egress)PE设备。所述第一转发设备和所述第二转发设备可处于同一转发路径上，该转发路径为用于转发所述第一业务流的路径。
可选地，所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之前，所述方法还包括：所述第一转发设备向所述第二转发设备发送控制信息，所述控制信息用于控制所述低延迟转发模式的状态。可选地，所述控制信息可对应所述第一业务流；所述低延迟转发模式是与所述第一业务流对应的转发模式。
其中，所述控制所述低延迟转发模式的状态包括控制所述低延迟转发模式的开启和关闭。所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，或者所述控制信息包括所述低延迟转发模式的开始时刻和运行时长。可选地，所述控制所述低延迟转发模式的状态包括控制所述低延迟转发模式的开启。 比如：所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式。这样，所述第一转发设备在处理所述第一业务流之前，向第二转发设备发送所述控制消息，省略了管理平面为复杂的配置操作，而且能够更灵活地实现对低延迟业务流的处理。
其中，若所述控制信息包括所述开启标识，则在所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之后，所述方法还包括：所述第一转发设备向所述第二转发设备发送关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式。
可选地，所述控制信息可携带于资源预留协议(Resource Reservation Protocol，RSVP)消息或通用关联通道头(generic associated channel header，G-ACH)通道消息。比如，所述控制信息可携带于所述RSVP消息或G-ACH通道消息包括的扩展类型长度值(type-length-value，TLV)中。
可选地，在所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之前，所述方法还包括：所述第一转发设备向所述第二转发设备发送检测报文，所述检测报文包括第一时延值和第二时延值，所述第一时延值为所述转发路径上的转发设备允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长。其中，所述检测报文用于获得所述第一业务流的转发路径上的转发设备产生的传输延迟。
可选地，所述第一转发设备向所述第二转发设备发送所述检测报文可在所述第一转发设备向所述第二转发设备发送所述控制信息之后执行。所述第二转发设备属于用于转发所述第一业务流的转发路径。这样，所述第一转发设备在所述转发路径包括的转发设备上设置了低延迟转发模式的开始时刻后，再发送所述检测报文，可以对设置了低延迟转发模式状态下的转发设备的传输时延进行检测，有助于利用检测结果定位传输延迟较大的转发设备，进一步降低端到端的传输时延。
可选地，所述第一转发设备向第二转发设备发送的控制信息还包括所述第一业务流的带宽需求。这样，所述第一转发设备可通过发送所述控制信息，使得所述第二转发设备根据所述第一业务流的带宽需求，预先分配相应的第二转发资源，无需再发送用于进行转发资源分配的消息。
可选地，所述低延迟标识可以是低延迟标签。比如对通常的用于转发的多协议标签交换(Multiprotocol Label Switching，MPLS)标签进行扩展，使其具有低延迟标识的功能和转发的功能。或者，在数据报文中增加在转发该数据报文的过程中不会被丢弃或替换的低延迟标签。
第二方面，提供了一种用于处理低延迟业务流的方法，包括：
第二转发设备接收来自第一转发设备的第二数据报文，所述第二数据报文包括低延迟标识，所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式，所述第二数据报文属于所述第一业务流；
所述第二转发设备根据所述低延迟标识，在低延迟转发模式下处理所述第二数据报文。
本申请实施例提供的方法中，第二转发设备可根据接收到的数据报文中是否包括低延迟标识来确定是否在低延迟模式下发送所述接收到的数据报文，无需通过管理平面进行复杂的配置操作，且低延迟模式的运行较为灵活。
其中，所述第二数据报文还包括所述第一数据报文，若所述第二转发设备是egress PE设备，则所述第二转发设备根据所述延迟标识，在低延迟转发模式下处理所述第二数据报文包括：所述第二转发设备删除所述第二数据报文中的低延迟标识，获得所述第一数据报文；所述第二转发设备根据所述延迟标识，在低延迟模式下，向与所述第二转发设备通信的CE设备发送所述第一数据报文。
其中，若所述第二转发设备是transit P设备，则所述第二转发设备根据所述延迟标识，在低延迟转发模式下处理所述第二数据报文包括：所述第二转发设备根据所述延迟标识，在低延迟模式下，向第三 转发设备发送所述第二数据报文。所述第三转发设备为转发路径上沿第一方向所述第二转发设备的下一跳。所述转发路径为用于转发所述第一业务流的路径，所述第一方向为从ingress PE到egress PE的方向。可选地，所述第三转发设备为另一transit P设备或egress PE设备。
可选地，所述第二转发设备接收来自第一转发设备的第二数据报文之前，所述方法还包括：所述第二转发设备接收来自所述第一转发设备的控制信息，所述控制信息用于控制所述低延迟转发模式的状态；所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态。
可选地，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态包括：所述第二转发设备根据所述低延迟转发模式的开始时刻和所述结束时刻，运行所述低延迟转发模式。
可选地，所述控制信息包括所述低延迟转发模式的开始时刻和运行时长，所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态包括：所述第二转发设备根据所述低延迟转发模式的开始时刻和所述运行时长，运行所述低延迟转发模式。
可选地，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式，所述第二转发设备根据所述低延迟标识，在低延迟转发模式下发送所述第二数据报文之后，所述方法还包括：所述第二转发设备接收所述第一转发设备发送的关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式；所述第二转发设备根据所述关闭标识，停止在所述低延迟转发模式发送所述第一业务流的数据报文。
其中，所述停止在所述低延迟转发模式发送所述第一业务流的数据报文可以是不对所述第一业务流的数据报文进行转发。所述停止在所述低延迟转发模式发送所述第一业务流的数据报文还可以是采用通常的方式对所述第一业务流的数据报文进行发送，即不再优先处理所述第一业务流的数据报文。
可选地，所述控制信息的发送方式与第一方面相同，在此不再赘述。
可选地，所述方法还包括：所述第二转发设备接收来自所述第一转发设备的第一检测报文，所述第一检测报文包括第一时延值和第二时延值，所述第一检测报文用于获得转发路径上的转发设备产生的传输延迟，所述第一时延值为所述转发路径上的转发设备可允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长；所述第二转发设备获得第三时延值，所述第三时延值为所述第二转发设备接收到所述第一检测报文到发送所述第一检测报文间的时长；所述第二转发设备根据所述第一检测报文和所述第三时延值，获得第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。
本申请实施例中的第二转发设备可根据接收到的来自第一转发设备的检测报文，比如第一检测报文，模拟数据报文的传输延迟，将转发所述第一检测报文产生的传输延迟添加至所述第一检测报文，获得第二检测报文。这样，接收到所述第二检测报文的设备，可利用所述第二检测报文中携带的信息和/或参数，定位传输延迟较大的转发设备，有助于进一步降低端到端的传输时延。
可选地，所述第二转发设备为transit P设备，所述方法还包括：所述第二转发设备向第三转发设备发送所述第二检测报文，所述第三转发设备为所述转发路径上所述第二转发设备的下一跳。
第三方面，提供了一种第一转发设备，所述第一转发设备包括：
处理单元，用于确定接收的第一数据报文属于第一业务流后，获得与所述第一业务流对应的低延迟标识，所述第一转发设备是作为网络入口的设备；
获得单元，用于根据所述第一数据报文和所述低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式；
第一发送单元，用于在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。
可选地，所述第一转发设备还包括：
第二发送单元，用于向所述第二转发设备发送控制信息，所述控制信息用于控制所述低延迟转发模式的状态。
其中，所述第二发送单元可在所述第一发送单元在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之前，向所述第二转发设备发送所述控制信息。这样，如果所述第一数据报文是所述第一业务流的首个数据报文，所述第一转发设备在发送包括所述第一数据报文的所述第二数据报文之前，将用于控制所述低延迟转发模式的状态的控制信息发送给所述第二转发设备即可，使得所述低延迟转发模式的配置更为灵活。
可选地，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，或者所述控制信息包括所述低延迟转发模式的开始时刻和运行时长。
可选地，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式，所述第一转发设备还包括：
第三发送单元，用于向所述第二转发设备发送关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式。
可选地，所述第一转发设备还包括：
第四发送单元，用于向所述第二转发设备发送检测报文，所述检测报文包括第一时延值和第二时延值，所述第一时延值为转发路径上的转发设备允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长。其中，所述检测报文用于获得所述转发路径上转发设备产生的传输延迟。所述转发路径包括所述第一转发设备和所述第二转发设备，所述转发路径为用于转发所述第一业务流的路径。
可选地，所述检测报文还包括所述第一转发设备发送所述检测报文的时刻。这样，作为网络出口的转发设备可根据所述第一转发设备发送所述检测报文的时刻、所述作为网络出口的转发设备接收到所述检测报文的时刻、所述检测报文中携带的转发设备产生的延迟值，确定转发设备产生的延迟值和转发路径上的物理链路产生的延迟值，有助于定位延迟值较大的转发设备和/或物理链路。
上述第三方面或第三方面的任意一种可能的实现方式提供的第一转发设备，可采用第一方面或第一方面的任意一种可能的实现方式提供的方法。
第四方面，提供了一种第二转发设备，所述第二转发设备包括：
第一接收单元，用于接收来自第一转发设备的第二数据报文，所述第二数据报文包括低延迟标识，所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式，所述第二数据报文属于所述第一业务流；
第一发送单元，用于根据所述低延迟标识，在所述低延迟转发模式下处理所述第二数据报文。
可选地，所述第二转发设备还包括：
第二接收单元，用于接收来自所述第一转发设备的控制信息，所述控制信息用于控制所述低延迟转发模式的状态；
控制单元，用于根据所述控制信息，动态控制所述低延迟转发模式的状态。
其中，所述第二接收单元在所述第一接收单元接收到所述第二数据报文之前接收到所述控制信息即可，这样所述控制单元可在所述第一接收单元接收到所述第二数据报文之前，完成对所述低延迟转发模式的状态的动态控制，比如开启所述低延迟转发模式。
可选地，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，所述控制单元具体用于根 据所述低延迟转发模式的开始时刻和所述结束时刻，运行所述低延迟转发模式。
可选地，所述控制信息包括所述低延迟转发模式的开始时刻和运行时长，所述控制单元具体用于根据所述低延迟转发模式的开始时刻和所述运行时长，运行所述低延迟转发模式。
可选地，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式，所述第二转发设备还包括：
第三接收单元，用于接收所述第一转发设备发送的关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式；
所述控制单元还用于根据所述关闭标识，停止在所述低延迟转发模式发送所述第一业务流的数据报文。
可选地，所述第二转发设备还包括：
第四接收单元，用于接收来自所述第一转发设备的第一检测报文，所述第一检测报文包括第一时延值和第二时延值，所述第一检测报文用于获得转发路径上转发设备产生的传输延迟，所述第一时延值为所述转发路径上的转发设备可允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长；
第一获得单元，用于获得第三时延值，所述第三时延值为所述第二转发设备接收到所述第一检测报文到发送所述第一检测报文间的时长；
第二获得单元，用于根据所述第一检测报文和所述第三时延值，获得第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。
可选地，所述第二转发设备为中间转发设备，所述第一发送单元具体用于向第三转发设备发送所述第二检测报文，所述第三转发设备为所述转发路径上沿第一方向所述第二转发设备的下一跳，所述第一方向为所述第一转发设备到作为网络出口的设备的方向。
上述第四方面或第四方面的任意一种可能的实现方式提供的第二转发设备，可采用第二方面或第二方面的任意一种可能的实现方式提供的方法。
第五方面，提供了一种第一转发设备，该第一转发设备包括：处理器、存储器和通信接口。所述处理器、所述存储器和所述通信接口通过通信总线连接。所述存储器用于存储程序。所述处理器根据从所述存储器中读取的程序所包括的可执行指令，执行上述第一方面或第一方面的任意一种可能的实现方式提供的方法。
可选地，所述第六方面提供的第一转发设备可以是所述第三方面提供的第一转发设备。
第六方面，提供了一种第二转发设备，该第二转发设备包括：处理器、存储器和通信接口。所述处理器、所述存储器和所述通信接口通过通信总线连接。所述存储器用于存储程序。所述处理器根据从所述存储器中读取的程序所包括的可执行指令，执行上述第二方面或第二方面的任意一种可能的实现方式提供的方法。
第七方面，提供了一种用于处理低延迟业务流的系统，该系统包括上述第三方面或第三方面的任意一种可能的实现方式提供的第一转发设备，和上述第四方面或第四方面的任意一种可能的实现方式提供的第二转发设备；或者该系统包括上述第五方面或第五方面的任意一种可能的实现方式提供的第一转发设备和上述第六方面或第六方面的任意一种可能的实现方式提供的第二转发设备。
附图说明
图1为一种网桥的某个端口的示意图。
图2为本申请实施例提供的一种网络场景示意图。
图3为本申请实施例一提供的用于配置低延迟业务转发模式的方法流程图。
图4为本申请实施例二提供的用于处理低延迟业务流的方法流程图。
图5(a)为本申请实施例提供的第一G-ACH通道消息的示意图。
图5(b)为本申请实施例提供的第二数据报文的示意图。
图5(c)为本申请实施例提供的第二数据报文的示意图。
图6为本申请实施例三提供的用于处理低延迟业务流的方法流程图。
图7为本申请实施例四提供的用于检测转发路径的传输时延的方法流程图。
图8为本申请实施例五提供的一种第一转发设备的示意图。
图9为本申请实施例六提供的一种第二转发设备的示意图。
图10为本申请实施例七提供的一种第一转发设备的示意图。
图11为本申请实施例八提供的一种第二转发设备的示意图。
图12为本申请实施例十提供的用于建立转发路径的方法流程图。
图13为本申请实施例提供的用于建立转发路径的网络场景示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚地描述。
图1为网桥的某个端口的示意图。如图1所示，网桥的某个端口包括了8个队列和8个门。其中，所述8个队列包括队列0、队列1、队列2、队列3、队列4、队列5、队列6和队列7。所述8个门包括门0、门1、门2、门3、门4、门5,、门6和门7。每个队列都对应一个发送选择算法。所述发送选择算法用来计算队列中待发送业务流的优先级并根据优先级输出业务流至相应的门。网桥的端口设置有门控列表，门控列表包括80个门控表项，比如T00、T01、T02…T78和T79。每个门控表项用于保存在各个指定时隙内的门控信号。门控信号的每一位用于控制相应的门开启或关闭，比如图1中的C表示关闭，图1中的O表示开启。如图1所示，在门控列表中的T04加载到所述8个门的情况下，T04包括的门控信号为OCOOCOOO，即门7开启，门6关闭，门5开启，门4开启，门3关闭，门2开启，门1开启和门0开启。队列7中高优先级的业务流通过发送选择算法选出后，队列7中的高优先级的业务流被输出至门7。由于门7根据T04的相应门控信号处于开启状态，队列7中的高优先级的业务流被输出。队列6中高优先级的业务流通过发送选择算法选出后，队列6中的高优先级的业务流被输出至门6。门6根据T04的相应门控信号处于关闭状态，队列6中的高优先级的业务流停止输出。网桥的某个端口对队列5、队列4…队列1和队列0中的业务流的处理方式与队列6和队列7相同，在此不再赘述。
通常为了满足低延迟业务的处理，通常会按照图1对网桥的某个端口进行配置。与低延迟业务有关的网桥的端口均需要通过管理平面配置一个门控列表，提高了管理的复杂性。由于门控列表包括的门控表项的数量有限，可能需要在所有门控表项被使用完之前根据业务流再进行相应的配置，灵活性较差。
针对上述问题，本申请提出了有助于降低管理的复杂性和提高控制的灵活性的方法。所述方法是作为网络入口的第一转发设备对接收到的第一数据报文进行判断，在确定所述第一数据报文属于第一业务流后，所述第一转发设备根据所述第一数据报文和与所述第一业务流对应的低延迟标识，获得第二数据报文。所述第一转发设备在低延迟转发模式下，向第二转发设备发送所述第二数据报文。所述第二转发设备根据所述第二数据报文包括的所述低延迟标识，在所述低延迟转发模式下发送所述第二数据报文。所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式。所述解决方法可通过以下几种实施例实现。
本申请实施例中的低延迟业务流为要求传输时延在预设的时长内的业务流。所述预设的时长通常较 低，比如毫秒级，可以是5毫秒级别或10毫秒级别。所述传输时延为网络中端到端的传输过程中产生的时延。在本申请中，所述网络中端到端的传输指代的是作为网络入口的设备到作为网络出口的设备间的传输，或者所述网络中端到端的传输指代的是报文的源地址到报文的目的地址间的传输。所述低延迟业务流可包括一个或多个IP分组，或者所述低延迟业务流可包括一个或多个以太帧。本申请实施例中的低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式，有助于保证低延迟业务流能够得到优先处理，比如采用时隙门控功能对相应的队列中的低延迟业务流进行优先处理。所述时隙门控功能是在预设的时间段内控制门开启或关闭的功能，所述门可用于传输与其对应队列中的业务流。所述优先处理包括优先调度和优先转发。
下面提到的“实施例一”至“实施例十”仅用来标识各个实施例，而不是标识一些实施例比另一些实施例更优。
图2为本申请实施例提供的一种网络场景示意图。下面结合图2，对本申请实施例提供的网络场景进行说明。CE1、CE2、CE3和CE4是能够与用户通信的设备。PE1能够与CE1、CE2和P1进行通信。P1能够与PE2进行通信。PE2能够与CE3和CE4进行通信。CE1向CE3发送第一业务流。图2中的实线表示用于转发所述第一业务流的路径。所述用于转发所述第一业务流的路径为第一转发路径。CE2向CE4发送第二业务流。图2中的虚线表示用于转发所述第二业务流的路径。所述用于转发所述第二业务流的路径为第二转发路径。所述第一业务流和所述第二业务流均属于低延迟业务流。PE1可以是所述第一转发路径的入口节点和所述第二转发路径的入口节点。PE2可以是所述第一转发路径的出口节点和所述第二转发路径的出口节点。图2中的设备PE1、PE2、P1为运营商网络中的设备。
图2中的任一CE可以是宽带接入客户端、企业的出口网关或者是数据中心(data center，DC)的出口网关。图2中的任一PE可以是路由器或分组传送网(Packet Transport Network，PTN)设备。图2中的任一P可以是路由器或PTN设备。图2中的任一CE与相邻的可通信的PE间的链路可以是以太网链路、无源光网络(passive optical network，PON)链路或x数字用户线(x digital subscriber line，xDSL)链路。比如CE1与PE1间的链路，CE2与PE1间的链路，CE3与PE2间的链路和CE4与PE2间的链路可以是以太网链路、PON链路或xDSL链路，在此不再对不同的链路所形成的组合方式进行举例说明。
实施例一
图3为本申请实施例一提供的配置低延迟业务转发模式的方法流程图。本申请实施例一是从转发路径上的设备配置低延迟转发模式的角度，对用于处理低延迟业务流的方法进行说明。图2的运营商网络中的所有设备需要预先部署时间同步协议，比如电气和电子工程师学会(Institute of Electrical and Electronics Engineers，IEEE)的1588时钟同步协议，即IEEE 1588v2，使得运营商网络中的所有设备都有全网同步的时间。可选地，图2的运营商网络中的所有设备还可预先部署频率同步协议，比如同步以太网(Synchronous Ethernet，SyncE)协议，使得运营商网络中的所有设备都有全网同步的频率协议。本申请实施例中运营商网络中的转发设备可在配置低延迟转发模式之前，采用上述时钟同步协议进行时间同步，还可采用上述频率同步协议进行频率同步。本申请实施例不再对运营商网络中的所有设备进行时钟同步和/或频率同步的方法进行说明。本申请实施例中的入口节点为用于转发某一业务流的转发路径在运营商网络中的入口，比如PE1。出口节点为所述用于转发某一业务流的转发路径在运营商网络中的出口，比如PE2。
本申请实施例一提供的方法包括第一低延迟转发模式的配置过程和第二低延迟转发模式的配置过程。下面结合图2和图3，对本申请实施例一提供的用于配置低延迟业务转发模式的方法进行说明。
301，PE1向P1发送第一控制信息。
举例说明，所述第一控制信息是用于控制第一低延迟转发模式的状态。所述第一低延迟转发模式为用于处理第一业务流的设备采用的低延迟转发模式，即所述第一低延迟转发模式与所述第一业务流对应。所述第一低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式。所述用于处理第一业务流的设备可以是图2的运营商网络中用于处理所述第一业务流的转发设备。对于实施例一的运营商网络来说，PE1、P1和PE2上设置有所述第一低延迟转发模式。所述第一低延迟转发模式的状态包括所述第一低延迟转发模式的开始和所述第一低延迟转发模式结束。或者所述第一低延迟转发模式的状态包括所述第一低延迟转发模式的开始。
举例说明，所述第一控制信息包括所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的结束时刻。或者所述第一控制信息包括所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的运行时长。或者所述第一控制信息包括第一开启标识。所述第一开启标识用于指示获得所述第一开启标识的设备开启所述第一低延迟转发模式。
举例说明，所述第一低延迟转发模式的开始时刻可表示为T1。所述第一低延迟转发模式的结束时刻可表示为T2。所述第一低延迟转发模式的运行时长可表示为t。T1和T2的获取方式可以为方式一到方式三中的任何一种。
方式一，CE1在接入认证的过程中，认证设备获得通过认证的用户(与CE1通信且通过认证的用户)的T1和T2，PE1从所述认证设备获得T1和T2。其中，T1可设置为所述用户通过认证的时刻。本申请实施例中的时刻指代某一个时间点。
方式二，用户通过运营商提供的页面购买第一业务，在用户购买成功后，PE1可从运营商提供的服务器获得T1和T2。
方式三，用户侧的与第一业务对应的应用程序获得T1和T2后，PE1从用户(与CE1通信的用户)获得T1和T2。PE1从用户获得T1和T2的方法可以是用户向PE1主动上报T1和T2。所述第一业务流是所述第一业务对应的数据流。T1和t的获取方式可与上述T1和T2的获取方式相同，在此不再赘述。
举例说明，PE1在建立第一转发路径的过程中发送所述第一控制信息，所述第一转发路径为用于转发所述第一业务流的路径。PE1建立第一转发路径的方法可参见实施例十提供的方法。P1为所述第一转发路径上的设备。所述第一控制信息可携带在资源预留协议(Resource Reservation Protocol，RSVP)消息或通用关联通道头(generic associated channel header，G-ACH)通道消息。
以携带所述第一控制信息的RSVP消息为例，可以对RSVP消息进行扩展，比如在RSVP消息中扩展一个低延迟字段，比如增加一个类型长度值(type-length-value，TLV)。所述低延迟字段携带的TLV可用来携带所述第一控制信息。可选地，在所述第一低延迟转发模式进一步包括子转发模式的情况下，所述低延迟字段携带的TLV还可用来携带第一控制标记(英文为flag)。所述第一控制标记用来标识所述子转发模式的开启或关闭。所述子转发模式可以是通常的抢占式转发(英文为preemption forwarding)或时隙调度(英文为timescale scheduling)。
以携带所述第一控制信息的第一G-ACH通道消息为例，可以对第一G-ACH通道消息进行扩展。比如第一G-ACH通道消息中增加关联通道头(associated channel header，ACH)TLV，ACH TLV可用于携带所述第一控制信息，如图5(a)所示。可在所述第一G-ACH通道消息外层依次封装通用ACH标签(generic associated channel label，GAL)、PW标签和第一标签。所述GAL用来指示存在G-ACH控制通道。其中，PW标签是可选的，例如对于某些二层业务来说需要添加PW标签，对于三层业务来说无需添加PW标签。本申请实施例中用于携带所述第一控制信息的第一G-ACH通道消息可采用图5(a)中的结构，在下述实施例中不再逐一举例说明。可选地，在所述第一低延迟转发模式进一步包括 子转发模式的情况下，所述ACH TLV还可用来携带第一控制flag。所述第一控制flag用来标识所述子转发模式的开启或关闭。所述子转发模式可以是通常的preemption forwarding或timescale scheduling。
302，PE1向PE2发送第二控制信息。
举例说明，第二控制信息是用于控制第二低延迟转发模式的状态。所述第二低延迟转发模式为用于处理第二业务流的设备采用的低延迟转发模式，即所述第二低延迟转发模式与所述第二业务流对应。所述第二低延迟转发模式是动态控制下实现所述第二业务流的快速转发的模式。所述用于处理第二业务流的设备可以是图2的运营商网络中用于处理所述第二业务流的转发设备。对于实施例一的运营商网络来说，PE1和PE2上还设置有所述第二低延迟转发模式。所述第二低延迟转发模式的状态包括所述第二低延迟转发模式的开始和结束。或者所述第二低延迟转发模式的状态包括所述第二低延迟转发模式的开始。所述第二低延迟转发模式的开始指代所述第二低延迟转发模式的开启。所述第二低延迟转发模式的结束指代所述第二低延迟转发模式的结束。
举例说明，第二控制信息包括所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的结束时刻。或者所述第二控制信息包括所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的运行时长。或者所述第二控制信息包括第二开启标识。所述第二开启标识用于指示获得所述第二开启标识的设备开启所述第二低延迟转发模式。
举例说明，PE1在建立第二转发路径的过程中发送所述第二控制信息。所述第二转发路径用于传输所述第二业务流。PE2为所述第二转发路径上的设备，也是所述第二转发路径上作为出口节点的设备。所述第二控制信息的发送方式与所述第一控制信息的发送方式相同，在此不再赘述。
以携带所述第二控制信息的RSVP消息为例，可以对所述RSVP消息进行扩展。其中，所述扩展所述RSVP消息以携带所述第二控制信息的方式与扩展所述RSVP消息以携带所述第一控制信息的方式相同，在此不再赘述。
以携带所述第二控制信息的第二G-ACH通道消息为例，可以对第二G-ACH通道消息进行扩展，比如所述第二G-ACH消息包括的ACH TLV可用于携带所述第二控制信息。其中，所述扩展所述第二G-ACH通道消息以携带所述第二控制信息的方式，与所述扩展所述第一G-ACH通道消息的方式相同，在此不再赘述。
303，P1根据所述第一控制信息进行配置。
举例说明，P1根据所述第一控制信息配置所述第一低延迟转发模式的状态。在一种实现方式中，P1可根据所述第一控制信息生成与所述第一业务流对应的表项。若所述第一控制信息包括所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的结束时刻，则所述与所述第一业务流对应的表项包括：所述第一低延迟标识、所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的结束时刻。若所述第一控制信息包括所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的运行时长，则所述与所述第一业务流对应的表项包括：所述第一低延迟标识、所述第一低延迟转发模式的开始时刻和所述第一低延迟转发模式的运行时长。所述第一低延迟标识可来自于所述第一控制信息，或者通过静态配置的方式获得的，或者通过来自PE1的其他消息获得。在另一种实现方式中，P1可根据所述第一控制信息生成与第一业务对应的表项。若所述与所述第一业务流对应的表项还包括所述第一业务流的标识，则所述与第一业务对应的表项还包括所述第一业务的标识，即用所述第一业务的标识替换所述与第一业务流对应的表项包括的所述第一业务流的标识，以获得所述与第一业务对应的表项。
304，P1向PE2发送所述第一控制信息。
举例说明，P1可沿所述第一转发路径，向PE2发送所述第一控制信息。
其中，301、302和304的执行顺序需保证301先于303和304执行，302的执行顺序不限。302可 以与301同时执行，或者302可以先于301执行。或者302可以与303同时执行，或者302可以在303之后执行。或者302可以与304同时执行，或者302可以在304之后执行。303可在304之后执行，或者303可与304同时执行。
305，PE2根据所述第一控制信息进行配置。
举例说明，PE2根据所述第一控制信息配置所述第一低延迟转发模式的状态。PE2配置所述第一低延迟转发模式的状态的方式与303中P1采用的配置方式相同，在此不再赘述。PE2可采用303中P1采用的方式，获得与所述第一业务流对应的表项或者获得与所述第一业务对应的表项。PE2获得的与所述第一业务流对应的表项可称为第一表项。PE2获得的与所述第一业务流对应的表项与303中P1获得与所述第一业务流对应的表项相同。或者PE2获得的与第一业务对应的表项可称为第一表项。PE2获得的与第一业务对应的表项与303中P1获得的与所述第一业务对应的表项相同。
306，PE2根据所述第二控制信息进行配置。
举例说明，PE2根据所述第二控制信息配置所述第二低延迟转发模式的状态。在一种实现方式中，PE2可根据所述第二控制信息生成第二表项，所述第二表项为与所述第二业务流对应的表项。若第二控制信息包括所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的结束时刻，则所述第二表项包括：所述第二低延迟标识、所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的结束时刻。若所述第二控制信息包括所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的运行时长，则所述第二表项包括：所述第二低延迟标识、所述第二低延迟转发模式的开始时刻和所述第二低延迟转发模式的运行时长。所述第二低延迟标识可来自于所述第二控制信息，或者通过静态配置的方式获得的，或者通过来自PE1的其他消息获得。在另一种实现方式中，PE2可根据所述第二控制信息生成与第二业务对应的表项，即所述与第二业务对应的表项为所述第二表项。若所述与所述第二业务流对应的表项还包括所述第二业务流的标识，则用所述第二业务的标识替换所述第二业务流的标识，以获得所述与第二业务对应的表项。
其中，305在304之后执行，306在302之后执行。305和306的先后顺序不限，比如305可以和306同时执行，或者305在306之后执行。
对于301中PE1向P1发送第一控制信息来说，可选地，所述第一控制信息还可包括第一低延迟(英文为low latency)标识。所述第一低延迟标识与所述第一业务流对应。所述第一低延迟标识用于指示接收到所述第一业务流的转发设备在第一低延迟转发模式下转发所述第一业务流。本申请实施例对于第一低延迟标识的具体形式不进行限定。这样，PE1可通过携带所述第一控制信息的消息发送所述第一低延迟标识，节省网络中报文的发送数量，有助于节省网络资源。可选地，PE1还可在发送所述第一控制信息之前、之后或同时，沿所述第一转发路径发送携带所述第一低延迟标识的消息。
对于301中PE1向P1发送第一控制信息来说，可选地，所述第一控制信息还包括所述第一业务流的带宽需求和所述第一业务流允许的时延值。所述第一业务流允许的时延值为所述第一转发路径上的一台转发设备允许的时延值。这样，接收到所述第一业务流的带宽需求和所述第一业务流允许的时延值的设备，可用所述第一业务流的带宽需求和所述第一业务流允许的时延值分配第一转发资源。所述第一转发资源用于处理所述第一业务流。可选地，PE1还可在发送所述第一控制信息之前、之后或同时，沿所述第一转发路径发送携带所述第一业务流的带宽需求和所述第一业务流允许的时延值的消息。
对于302中PE1向PE2发送第二控制信息来说，可选地，所述第二控制信息还可包括第二低延迟标识。所述第二低延迟标识与所述第二业务流对应。所述第二低延迟标识用于指示接收到所述第二业务流的转发设备在第二低延迟转发模式下转发所述第二业务流。本申请实施例对于第二低延迟标识的具体形式不进行限定。这样，PE1可通过携带所述第二控制信息的消息发送所述第二低延迟标识，节省网络 中报文的发送数量，有助于节省网络资源。可选地，PE1还可在发送所述第二控制信息之前、之后或同时，沿所述第二转发路径发送携带所述第二低延迟标识的消息。
对于302中PE1向PE2发送第二控制信息来说，可选地，所述第二控制信息还包括所述第二业务流的带宽需求和所述第二业务流允许的时延值。所述第二业务流允许的时延值为所述第二转发路径上的一台转发设备允许的时延值。这样，接收到所述第二业务流的带宽需求和所述第二业务流允许的时延值的设备，可用所述第二业务流的带宽需求和所述第二业务流允许的时延值分配第二转发资源。所述第二转发资源用于处理所述第二业务流。可选地，PE1还可在发送所述第二控制信息之前、之后或同时，沿所述第二转发路径发送携带所述第二业务流的带宽需求和所述第二业务流允许的时延值的消息。
对于303中P1根据所述第一控制信息进行配置来说，可选地，P1可根据所述第一控制信息包括的所述第一业务流的带宽需求和所述第一业务流允许的时延值，分配用于转发所述第一业务流的第一转发资源。例如：P1可为所述第一业务流分配与所述第一业务流匹配的抢夺优先队列，即为所述第一业务流分配与所述第一业务流匹配的漏桶(英文为bucket)。其中，与所述第一业务流匹配的漏桶为所述第一业务流的带宽需求和所述第一业务流允许的时延值的乘积。这样，有助于实现所述第一业务流在P1产生的实际时延小于所述第一业务流允许的时延值。
对于303中P1根据所述第一控制信息进行配置来说，可选地，P1还可为所述第一业务流分配与所述第一业务流匹配的信用增长率(英文为credit rate)。相对于所述与所述第一业务流匹配的bucket来说，所述credit rate表示所述第一业务流允许的时延值内能够发送的数据分组的字节总数。
对305中PE2根据所述第一控制信息进行配置来说，可选地，PE2分配用于转发所述第一业务流的第一转发资源的方式与303中P1采用的分配方式相同，在此不再赘述。
对305中PE2根据所述第一控制信息进行配置来说，可选地，PE2还可为所述第一业务流分配与所述第一业务流匹配的credit rate，具体分配方式与303中P1采用的分配方式相同，在此不再赘述。
对306中PE2根据所述第二控制信息进行配置来说，可选地，PE2分配用于转发所述第二业务流的第二转发资源的方式与303中P1分配用于转发所述第一业务流的第一转发资源的方式相同，在此不再赘述。其中，与所述第二业务流匹配的漏桶为所述第二业务流的带宽需求和所述第二业务流允许的时延值的乘积，这样，有助于实现所述第二业务流在PE2产生的实际时延小于所述第二业务流允许的时延值。
对306中PE2根据所述第二控制信息进行配置来说，可选地，PE2还可为所述第二业务流分配与所述第二业务流匹配的credit rate。相对于所述与所述第二业务流匹配的bucket来说，所述credit rate表示所述第二业务流允许的时延值内能够发送的数据分组的字节总数。
可选地，本申请实施例一提供的方法还包括：PE1根据所述第一控制信息进行所述第一低延迟转发模式的配置。PE1在处理所述第一业务流之前，根据所述第一控制信息进行所述第一低延迟转发模式的配置。PE1进行所述第一低延迟转发模式的配置方法与303中P1采用的配置方法相同，在此不再赘述。
可选地，本申请实施例一提供的方法还包括：PE1根据所述第二控制信息进行所述第二低延迟转发模式的配置。PE1在处理所述第二业务流之前，根据所述第二控制信息进行所述第二低延迟转发模式的配置。PE1进行所述第二低延迟转发模式的配置方法与306中PE2采用的配置方法相同，在此不再赘述。
可选地，本申请实施例中的所述第一低延迟标识还用于指示属于所述第一业务流的数据报文。所述第二低延迟标识还用于指示属于所述第二业务流的数据报文。
其中，本申请实施例中，第一低延迟转发模式的配置过程和第二低延迟转发模式的配置过程是相互独立的过程。
本申请实施例一提供的方法中，作为网络入口的设备，比如PE1，向第一转发路径上的设备发送第一控制信息，以便接收到所述第一控制信息的设备，比如P1或PE2，根据所述第一控制信息配置第一低延迟转发模式的状态，以完成第一低延迟转发模式的配置。PE1还可向第二转发路径上的设备发送第二控制信息，以便接收到所述第二控制信息的设备，比如PE2，根据所述第二控制信息配置第二低延迟转发模式的状态，以完成第二低延迟转发模式的配置。本申请实施例一提供的方法中，作为网络入口的设备可通过第一控制信息和/或第二控制信息的发送，对转发路径上的设备的低延迟转发模式进行动态控制。所述第一转发路径和/或所述第二转发路径上的设备无需通过管理平面配置与业务流对应的门控列表，也无需进行门控列表的维护和更新操作，有助于降低管理面的复杂性和提高控制的灵活性。
实施例二
图4为本申请实施例二提供的用于处理低延迟业务流的方法流程图。本申请实施例二是以运营商网络为多协议标签交换(Multiprotocol Label Switching，MPLS)网络，低延迟业务流为第一业务流，对用于处理低延迟业务流的方法进行说明。图2中的PE1、P1和PE2上运行MPLS协议。用于转发所述第一业务流的第一转发路径为第一标签交换路径(label switched path，LSP)。与所述第一业务流对应的第一低延迟标识为第一低延迟标签。PE1、P1和PE2在处理第一业务流之前，完成第一低延迟转发模式的配置，具体可参见实施例一提供的方法。实施例一提供的方法可穿插进实施例二中以形成另一实施例，对于实施例一穿插进实施例二中形成的实施例不再进行说明。PE1、P1和PE2上可分别配置有与所述第一LSP对应的标签，所述与所述第一LSP对应的标签用来实现报文沿所述第一LSP进行转发，具体可参见IETF RFC3209，本申请实施例二中不再对与所述第一LSP对应的标签的分发过程进行说明。下面结合图2和图4，对本申请实施例二提供的用于处理低延迟业务流的方法进行说明。
401，PE1接收第一数据报文。
举例说明，PE1通过第一端口接收第一数据报文。所述第一数据报文可来自CE1。所述第一数据报文包括第一MAC地址和第一IP地址。所述第一MAC地址是CE1的MAC地址。所述第一IP地址是CE1的IP地址。所述第一数据报文还可包括第二MAC地址和第二IP地址，所述第二MAC地址是CE3的MAC地址。所述第二IP地址是CE3的IP地址。CE1和CE3可属于同一虚拟专用网(virtual private network，VPN)。所述第一数据报文可以是IP分组或以太帧。
可选地，401和402之间，本申请实施例一提供的方法还包括：PE1判断所述第一数据报文是否属于第一业务流；如果所述第一数据报文属于所述第一业务流，PE1执行402。如果所述第一数据报文不属于所述第一业务流，PE1将所述第一数据报文放入待转发队列，按照通常的调度方法对待转发队列中的报文进行处理。其中，通常的调度方法可以是按照入队的先后顺序出对，本申请实施例对于通常的调度方法不再进行逐一举例说明。
举例说明，PE1判断所述第一数据报文是否属于所述第一业务流的方式可以采用方式一至方式四中的任意一种方式。
第一种方式：多元组可包括所述第一端口、所述第一MAC地址、所述第一IP地址、所述第二MAC地址和所述第二IP地址、服务类别(class of service，CoS)、信流类型(TC，Traffic Class)和传输控制协议(Transmission Control Protocol，TCP)的端口号中的一个或多个。PE1可根据所述多元组，判断所述第一数据报文是否属于所述第一业务流。PE1上可保存有所述第一业务流的特征信息表。所述特征信息表中可包括所述第一业务流的标识和属于所述第一业务流的数据报文对应的特征信息。所述特征信息包括MAC地址、IP地址、应用层端口号、VLAN信息、VXLAN信息和物理层端口号中的一个或多个。其中，VLAN信息可以包括VLAN标识(identifier，ID)和/或优先级。VXLAN信息可以包括VXLAN网络标识(VXLAN Network Identifier，VNI)。PE1用所述多元组查询所述第一业务流的特征 信息表。若所述第一业务流的特征信息表存在与所述多元组匹配的信息，则PE1确定所述第一数据报文属于所述第一业务流。所述匹配指代所述第一业务流的特征信息表中存在与所述多元组相同的信息。比如，所述多元组包括N个信息，所述N为大于或等于1整数，所述第一业务流的特征信息表中存在与所述N个信息分别相同的信息，则PE1可确定所述第一数据报文属于所述第一业务流。
第二种方式：PE1上保存的所述特征信息表包括第一业务的标识和属于所述第一业务的数据报文对应的特征信息。所述属于所述第一业务的数据报文对应的特征信息可以与第一种方式中所述属于所述第一业务流的数据报文对应的特征信息相同。PE1根据所述属于所述第一业务的数据报文对应的特征信息，判断所述第一数据报文是否属于所述第一业务流的方法与第一种方式相同。如果PE1收到的第一数据报文包含属于所述第一业务流的数据报文对应的特征信息，则PE1确定该收到的第一数据报文属于所述第一业务流。
第三种方式：所述第一数据报文还包括所述第一数据报文所属的业务流的标识。若所述第一数据报文所属的业务流的标识为所述第一业务流的标识，则PE1确定所述第一数据报文属于所述第一业务流。
第四种方式：所述第一数据报文还包括所述第一数据报文所属的业务的标识。若所述第一数据报文所属的业务的标识为所述第一业务的标识，则PE1确定所述第一数据报文属于所述第一业务，即所述第一数据报文属于所述第一业务流。所述第一业务流为所述第一业务的数据流。
402，PE1确定所述第一数据报文属于第一业务流。
举例说明，PE1可在确定所述第一数据报文属于所述第一业务流后，获得第一低延迟标签。PE1获取所述第一低延迟标签的方式可为方式一至方式三中的任意一种。
方式一：PE1可根据401中的特征信息表，在确定所述第一数据报文属于所述第一业务流后，获得所述第一业务流的标识。或者所述第一数据报文中携带有所述第一业务流的标识，PE1可从所述第一数据报文中获得所述第一业务流的标识。PE1上配置有第一低延迟标签和所述第一业务流的标识的对应关系。PE1可根据所述对应关系和所述第一业务流的标识，获得所述第一低延迟标签。
方式二：若401中的特征信息表包括所述第一业务的标识，PE1可在确定所述第一数据报文属于所述第一业务流后，获得所述第一业务的标识。或者所述第一数据报文中携带有所述第一业务的标识，PE1可从所述第一数据报文中获得所述第一业务的标识。PE1上配置有所述第一低延迟标签和所述第一业务的标识的对应关系。PE1可根据所述对应关系和所述第一业务的标识，获得所述第一低延迟标签。
方式三：若PE1上只配置有所述第一低延迟标签，且PE1只用于转发所述第一业务流，则PE1确定所述第一数据报文属于所述第一业务流后，直接获得所述第一低延迟标签。
403，PE1根据所述第一数据报文和第一低延迟标签，获得第二数据报文。
举例说明，所述第二数据报文包括所述第一数据报文。具体地，若所述第一业务为低延迟IP业务，则所述第二数据报文还包括所述第一低延迟标签和第一标签，所述第一低延迟标签和所述第一标签封装于所述第一数据报文外层。所述第一标签为PE1被分配的与所述第一LSP对应的标签。所述第一标签用于指示PE1沿所述第一LSP转发数据报文。若所述第一业务为多个数据中心(data center，DC)之间的低延迟业务，则所述第二数据报文还包括所述第一低延迟标签和第一段路由(segment routing，SR)标签，所述第一低延迟标签和所述第一SR标签封装于所述第一数据报文外层。在实施例二中，所述第一SR标签用于标识PE1和P1之间的链路。所述第一标签和所述第一SR标签是两种标签形式，所述第一标签和所述第一SR标签均是用来指示PE1转发数据报文。
可选地，若所述第一业务为低延迟以太网业务，则所述第二数据报文还可以包括ACH头和/或伪线(pseudo wire，PW)标签。例如对于某些二层业务来说需要添加PW标签，对于三层业务来说无需添加PW标签。所述PW标签用于实现二层以太业务。图5(b)和图5(c)为所述第二数据报文所采用 的可能的封装格式示意图。如图5(b)所示，所述第一低延迟标签可封装于PW标签与所述第一标签之间。或者如图5(c)所示，所述第一低延迟标签可封装于ACH头和PW标签之间。本申请实施例在不影响所述第二数据报文转发的前提下，对于所述第一低延迟标签可能的位置不进行限定。
其中，PE1获得所述第二数据报文后，PE1会根据所述第二数据报文携带的第一低延迟标签，将所述第二数据报文送至PE1上的待发送队列中，所述PE1上的待发送队列为PE1上与所述第一业务流对应的优先级队列。
可选地，所述第一低延迟标签还可用于指示所述第二数据报文属于所述第一业务流。
404，PE1在第一低延迟转发模式下，向P1发送所述第二数据报文。
在404之前，PE1可采用实施例一提供的方法配置所述第一低延迟转发模式，这样能够保证PE1在发送所述二数据报文之前或同时，PE1已开始运行所述第一低延迟转发模式，即PE1上用于转发所述第一业务流的门控处于开启状态。PE1可无需通过管理平面配置通常的门控列表，PE1可根据所述第一低延迟转发模式的开启或关闭，来实现低延迟业务的快速处理，控制较为灵活，管理较为简单。
举例说明，PE1在所述第一低延迟转发模式下，向P1发送所述第二数据报文包括：PE1在所述第一低延迟转发模式下，对PE1上的待发送队列中的报文进行识别，获得所述第二数据报文；PE1根据所述第二数据报文包括的所述第一低延迟标签，优先选择所述第二数据报文进行发送。可选地，若PE1采用了credit整形算法，则PE1优先选择所述第二数据报文进行发送之后，还包括：PE1降低credit的数值。在一种实现方式中，所述PE1上的待转发队里中可能还包括多个属于所述第一业务流的数据报文，PE1可继续优先选择所述多个属于所述第一业务流的数据报文进行发送，直至所述PE1上的待转发队里中的属于所述第一业务流的数据报文被发送完或者credit的数值降为0。
405，P1根据来自PE1的所述第二数据报文，获得第三数据报文。
举例说明，P1根据来自PE1的所述第二数据报文，获得第三数据报文包括：P1接收到PE1发送的所述第二数据报文；P1根据所述第二数据报文，获得第三数据报文。其中，若所述第二数据报文包括所述第一标签，则P1用第二标签替换所述第二数据报文包括的所述第一标签，获得所述第三数据报文。所述第二标签是P1被分配的与所述第一LSP对应的标签。所述第二标签用于指示P1沿所述第一转发路径发送所述第三数据报文。若第二SR标签用于标识P1与PE2之间的链路，且所述第二数据报文包括所述第一SR标签和所述第二SR标签，所述第一SR标签位于所述第二数据报文包括的标签栈的栈顶，则P1可弹出所述第一SR标签，获得所述第三数据报文。即所述第二SR标签位于所述第三数据报文包括的标签栈的栈顶。可选地，若所述第一SR标签还能标识P1与PE2之间的链路，则P1不对所述第二数据报文包括的第一SR标签进行处理，将所述第二数据报文作为所述第三数据报文。其中，所述第二标签和所述第二SR标签为两种标签形式。所述第二标签和所述第二SR标签均是用于指示P1转发数据报文。
其中，P1获得所述第三数据报文后，P1会根据所述第一低延迟标签，将所述第三数据报文送至P1上的待发送队列中，所述P1上的待发送队列为P1上与所述第一业务流对应的优先级队列。
406，P1在所述第一低延迟转发模式下，向PE2发送所述第三数据报文。
在406之前，P1可采用实施例一提供的方法配置所述第一低延迟转发模式，这样能够保证P1在发送所述三数据报文之前或同时，P1已开始运行所述第一低延迟转发模式，即P1上用于转发所述第一业务流的门控处于开启状态。P1可无需通过管理平面配置通常的门控列表，P1可根据所述第一低延迟转发模式的开启或关闭，来实现低延迟业务的快速处理，控制较为灵活，管理较为简单。
举例说明，P1在所述第一低延迟转发模式下，向PE2发送所述第三数据报文包括：P1在所述第一低延迟转发模式下，对P1上的待发送队列中的报文进行识别，获得所述第三数据报文；P1根据所述第 三数据报文包括的所述第一低延迟标签，优先选择所述第三数据报文进行发送。可选地，若P1采用了credit整形算法，则P1优先选择所述第三数据报文进行发送之后，还包括：P1降低credit的数值。在一种实现方式中，所述P1上的待转发队里中可能还包括多个属于所述第一业务流的数据报文，P1可继续优先选择所述多个属于所述第一业务流的数据报文进行发送，直至所述P1上的待转发队里中的属于所述第一业务流的数据报文被发送完或者credit的数值降为0。
407，PE2根据来自P1的所述第三数据报文，获得所述第一数据报文。
举例说明，PE2根据来自P1的所述第三数据报文，获得所述第一数据报文包括：PE2接收到P1发送的所述第三数据报文；PE2从所述第三数据报文中删除所述第一标签和所述第一低延迟标签，获得所述第一数据报文。PE2获得所述第一数据报文后，PE2会根据所述第三数据报文包括的所述第一低延迟标签，将所述第一数据报文送至PE2上的待发送队列中，所述PE2上的待发送队列为PE2上与所述第一业务流对应的优先级队列。若所述第三数据报文中还包括PW标签和/或ACH头，则PE2根据所述第三数据报文获得所述第一数据报文的过程中还会删除PW标签和/或ACH头。
408，PE2在所述第一低延迟转发模式下，向CE3发送所述第一数据报文。
在408之前，PE2可采用实施例一提供的方法配置所述第一低延迟转发模式，这样能够保证PE2在发送所述一数据报文之前或同时，PE2已开始运行所述第一低延迟转发模式，即PE2上用于转发所述第一业务流的门控处于开启状态。PE2可无需通过管理平面配置通常的门控列表，PE2可根据第一低延迟转发模式的开启或关闭，来实现低延迟业务的快速处理，控制较为灵活，管理较为简单。
举例说明，PE2在所述第一低延迟转发模式下，向CE3发送所述第一数据报文包括：PE2在所述第一低延迟转发模式下，优先选择PE2上的待发送队列中的所述第一数据报文进行发送，即优先将所述第一数据报文发送至CE3。可选地，若PE2采用了credit整形算法，则PE2优先选择所述第一数据报文进行发送之后，还包括：PE2降低credit的数值。在一种实现方式中，所述PE2上的待转发队里中可能还包括多个属于所述第一业务流的数据报文，PE2可继续优先选择所述多个属于所述第一业务流的数据报文进行发送，直至所述PE2上的待转发队里中的属于所述第一业务流的数据报文被发送完或者credit的数值降为0。
若实施例一中PE1发送的第一控制信息包括第一开启标识，所述第一开启标识用于标识开启所述第一低延迟转发模式，则本申请实施例二提供的方法还包括：
在401之前，PE1可采用实施例一提供的方法，向P1发送包括第一开启标识的第一控制信息。所述包括第一开启标识的第一控制信息可参见实施例一的相应内容，在此不再赘述。
在404之后，PE1可采用实施例一提供的方法，向P1发送第一关闭标识。所述第一关闭标识用于指示接收到所述第一关闭标识的转发设备停止在所述第一低延迟转发模式下处理所述第一业务流的数据报文。
409，PE1停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。
其中，本申请实施例可通过多种方式触发PE1停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。方式一，PE1可根据所述第一关闭标识，停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。方式二，作为入口节点的PE1可在确定完成所述第一业务流的最后一个数据报文的发送后，停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。方式三，作为入口节点的PE1可在其他控制设备或管理设备的控制下，停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。方式四，作为入口节点的PE1可在预定时长内未接收到所述第一业务流的数据报文后，停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。
410，P1根据PE1发送的第一关闭标识，停止在所述第一低延迟转发模式下发送所述第一业务流的 数据报文。
411，PE2根据P1发送的第一关闭标识，停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文。
409、410和411可在实施例二提供的方法所包括的任意步骤之间执行，409、410和411的执行取决于关闭标识的发送时刻。其中，所述停止在所述第一低延迟转发模式下发送所述第一业务流的数据报文可采用下述方式一或方式二。
方式一：关闭所述第一低延迟转发模式，且将在关闭所述第一低延迟转发模式后获得的所述第一业务流的数据报文送至低优先级的队列中。所述低优先级的队列和所述高优先级的队列是个相对的概念，所述低优先级的队列是相对于所述第一低延迟转发模式下的优先级队列而言的普通队列。
方式二：关闭所述第一低延迟转发模式，且丢弃在关闭所述第一低延迟转发模式后获得的所述第一业务流的数据报文。所述关闭所述第一低延迟转发模式可以是获得所述关闭标识后执行。
本申请实施例二提供的方法中，PE1可对P1和PE2上与所述第一业务流相关的门控进行灵活地控制，即发送用于控制所述第一低延迟转发模式的第一控制信息，无需通过管理平面为每个端口配置门控列表，降低管理的复杂性。
在图2所示的运营商网络为MPLS网络的情况下，PE1沿第二LSP，向PE2发送属于第二业务流的数据报文的方法，与上述PE1沿所述第一LSP，经过P1，向PE2发送属于所述第一业务流的数据报文的方法相似，只是无需经过P1的处理。所述第二LSP为图2中运营商网络为MPLS网络时虚线所表示的路径。
本申请实施例二是将所述第一低延迟标签和所述第一LSP对应的标签进行了区分，即所述第一LSP上的每个转发设备被分配的用于转发的标签和所述第一低延迟标签是不同的标签。在其他可能的实现方式中，可以对所述第一LSP上的每个转发设备被分配的用于指导转发的标签，比如SR标签、段(segment)标签、第一标签或第二标签进行扩展，使其既具有指示如何转发所述第一业务流(即指示如何转发属于所述第一业务流的数据报文)的功能，又具有所述第一低延迟标签的功能。例如：本申请实施例可以对所述第一标签和所述第二标签进行扩展，使得所述第一标签和所述第二标签还具有所述第一低延迟标签的功能。这样，PE1可无需在所述第一数据报文中添加所述第一低延迟标签，P1可根据所述第二数据报文中的第一标签确定采用所述第一低延迟转发模式处理所述第二数据报文，PE2可根据所述第三数据报文中的第二标签确定采用所述第一低延迟转发模式处理所述第三数据报文，有助于进一步降低MPLS网络中的转发设备的配置的复杂性。
实施例三
图6为本申请实施例三提供的用于处理低延迟业务流的方法流程图。本申请实施例三是以运营商网络为IP网络，低延迟业务流为第二业务流，对用于处理低延迟业务流的方法进行说明。图2中的PE1和PE2运行互联网协议(Internet Protocol，IP)。PE1和PE2在处理第二业务流之前，完成第二低延迟转发模式的配置。下面仅以实施例一执行完之后的实施例三为例进行说明，对于实施例一穿插进实施例三中形成的实施例不再进行说明。PE1和PE2间的第二转发路径可根据IP路由协议进行预先规划，例如中间系统到中间系统(Intermediate System to Intermediate System，IS-IS)流量工程(traffic engineering，TE)或者开放式最短路径优先流量工程(Open Shortest Path First-Traffic Engineering，OSPF-TE)，本申请实施例三中不再对第二转发路径的规划过程进行说明。下面结合图2和图6，对本申请实施例三提供的用于处理低延迟业务流的方法进行说明。
601，PE1接收第一数据报文。
举例说明，PE1通过第二端口接收第一数据报文。所述第一数据报文可来自于CE2。所述第一数据 报文包括第三MAC地址和第三IP地址。所述第三MAC地址是CE2的MAC地址。所述第三IP地址是CE2的IP地址。所述第一数据报文还可包括第四MAC地址和第四IP地址。所述第四MAC地址是CE4的MAC地址。所述第四IP地址是CE4的IP地址。CE2和CE4可属于同一VPN。所述第一数据报文可以是IP分组或以太帧。
可选地，601和602之间，本申请实施例三提供的方法还包括：PE1判断所述第一数据报文是否属于第二业务流；如果所述第一数据报文属于所述第二业务流，PE1执行602。如果所述第一数据报文不属于所述第二业务流，PE1将所述第一数据报文放入待转发队列，按照通常的调度方法对待转发队列中的报文进行处理。本申请实施例中的通常的调度方法与实施例二中的通常的调度方法相同，在此不再赘述。
举例说明，PE1可采用实施例二的401中的方法，判断所述第一数据报文是否属于第二业务流，具体可采用第一种方式至第四种方式中的任意一种方式。
第一种方式：多元组可包括所述第二端口、所述第三MAC地址、所述第三IP地址、所述第四MAC地址和所述第四IP地址中的一个或多个。PE1可根据所述多元组判断所述第一数据报文是否属于所述第二业务流。PE1上可保存有特征信息表，所述特征信息表中可包括所述第二业务流的标识和属于所述第二业务流的数据报文对应的特征信息。所述特征信息包括MAC地址、IP地址、应用层端口号、VLAN信息、VXLAN信息和物理端口中的一个或多个。PE1用所述多元组查询所述特征信息表。若所述特征信息表中存在与所述多元组匹配的信息，则PE1确定所述第一数据报文属于所述第二业务流。本申请实施例三中的匹配与实施例二中的匹配的含义相同，在此不再赘述。
第二种方式：PE1上保存的所述特征信息表包括第二业务的标识和属于所述第二业务的数据报文对应的特征信息。所述属于所述第二业务的数据报文对应的特征信息可以与第一种方式中所述属于所述第二业务流的数据报文对应的特征信息相同。PE1根据所述属于所述第二业务的数据报文对应的特征信息，判断所述第一数据报文是否属于所述第二业务流的方法与第一种方式相同，在此不再对具体的判断方法进行赘述。
第三种方式：所述第一数据报文还包括所述第一数据报文所属的业务流的标识，若所述第一数据报文所属的业务流的标识为所述第二业务流的标识，则PE1确定所述第一数据报文属于所述第二业务流。
第四种方式：所述第一数据报文还包括所述第一数据报文所属的业务的标识，若所述第一数据报文所属的业务的标识为所述第二业务的标识，则PE1确定所述第一数据报文属于所述第二业务，即所述第一数据报文属于所述第二业务流。
602，PE1确定所述第一数据报文属于第二业务流。
举例说明，PE1可在确定所述第一数据报文属于所述第二业务流后，获得第二低延迟标识。PE1获取所述第二低延迟标识的方式可采用方式一至方式三中的任意一种方式。
方式一，PE1可根据601中的特征信息表，在确定所述第一数据报文属于所述第二业务流后，获得所述第二业务流的标识。或者所述第一数据报文中携带有所述第二业务流的标识，PE1可从所述第一数据报文中获得所述第二业务流的标识。PE1上配置有第二低延迟标识和所述第二业务流的标识的对应关系。PE1可根据所述对应关系和所述第二业务流的标识，获得所述第二低延迟标识。
方式二，若601中的特征信息表包括所述第二业务的标识，PE1可在确定所述第一数据报文属于所述第二业务流后，获得所述第二业务的标识。或者所述第一数据报文中携带有所述第二业务的标识，PE1可从所述第一数据报文中获得所述第二业务的标识。PE1上配置有所述第二低延迟标识和所述第二业务的标识的对应关系。PE1可根据所述对应关系和所述第二业务的标识，获得所述第二低延迟标识。
方式三，若PE1上只配置有所述第二低延迟标识，且PE1只用于转发所述第二业务流，则PE1确 定所述第一数据报文属于所述第二业务流后，直接获得所述第二低延迟标识。
603，PE1根据所述第一数据报文和第二低延迟标识，获得第二数据报文。
举例说明，所述第二数据报文包括所述第一数据报文和隧道封装。所述隧道封装可封装于所述第一数据报文外层。所述隧道封装包括所述第二低延迟标识。所述隧道封装可以是通用路由封装协议(Generic Routing Encapsulation，GRE)封装。
其中，PE1获得所述第二数据报文后，PE1会根据所述第二低延迟标识，将所述第二数据报文送至PE1上的待发送队列中，所述PE1上的待发送队列为PE1上与所述第二业务流对应的优先级队列。
604，PE1在第二低延迟转发模式下，向PE2发送所述第二数据报文。
在604之前，PE1可采用实施例一提供的方法配置所述第二低延迟转发模式。这样，能够保证PE1在发送所述二数据报文之前，PE1已开始运行所述第二低延迟转发模式，即PE1上用于转发所述第二业务流的门控处于开启状态。PE1可无需通过管理平面配置通常的门控列表，可根据第二低延迟转发模式的开启或关闭，来实现低延迟业务的快速处理，控制较为灵活，管理较为简单。
举例说明，PE1在所述第二低延迟转发模式下，向PE2发送所述第二数据报文包括：PE1在所述第二低延迟转发模式下，对待发送队列中的报文进行识别，获得所述第二数据报文；PE1根据所述第二数据报文包括的所述第二低延迟标识，优先选择所述第二数据报文进行发送。其中，PE1可沿PE1与PE2之间的隧道，向PE2发送所述第二数据报文。
605，PE2根据来自PE1的所述第二数据报文，获得所述第一数据报文。
举例说明，PE2根据来自PE1的所述第二数据报文，获得所述第一数据报文包括：PE2接收到PE1发送的所述第二数据报文；PE2从所述第二数据报文中删除所述隧道封装，获得所述第一数据报文。PE2获得所述第一数据报文后，PE2会根据所述第二数据报文包括的所述第二低延迟标识，将所述第一数据报文送至PE2上的待发送队列中，所述PE2上的待发送队列为PE2上与所述第二业务流对应的优先级队列。
606，PE2在所述第二低延迟转发模式下，向CE4发送所述第一数据报文。
在606之前，PE2可采用实施例一提供的方法配置所述第二低延迟转发模式。这样，能够保证PE2在发送所述一数据报文之前，PE2已开始运行所述第二低延迟转发模式，即PE2上用于转发所述第二业务流的门控处于开启状态。PE2可无需通过管理平面配置通常的门控列表，可根据第二低延迟转发模式的开启或关闭，来实现低延迟业务的快速处理，控制较为灵活，管理较为简单。
举例说明，PE2在所述第二低延迟转发模式下，向CE4发送所述第一数据报文包括：PE2在所述第二低延迟转发模式下，优先选择PE2上待发送队列中的所述第一数据报文进行发送，即优先将所述第一数据报文发送至CE4。
若实施例三中PE1发送至PE2的第二控制信息包括第二开启标识，所述第二开启标识用于标识开启所述第二低延迟转发模式，则本申请实施例三提供的方法还包括：
在601之前，PE1可采用实施例一提供的方法，向P1发送包括第二开启标识的第二控制信息。所述包括第二开启标识的第二控制信息可参见实施例一的相应内容，在此不再赘述。
在604之后，PE1可采用实施例一提供的方法，向PE2发送第二关闭标识。所述第二关闭标识用于指示接收到所述第二关闭标识的转发设备停止在所述第二低延迟转发模式下处理所述第二业务流的数据报文。
607，PE1停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文。
其中，本申请实施例可通过多种方式触发PE1停止在所述第一低延迟转发模式下发送所述第二业务流的数据报文。方式一，PE1可根据所述第二关闭标识，停止在所述第二低延迟转发模式下发送所述 第二业务流的数据报文。方式二，作为入口节点的PE1可在确定完成所述第二业务流的最后一个数据报文的发送后，停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文。方式三，作为入口节点的PE1可在其他控制设备或管理设备的控制下，停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文。方式四，作为入口节点的PE1可在预定时长内未接收到所述第二业务流的数据报文后，停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文。
608，PE2根据接收到的来自PE1的第二关闭标识，停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文。
607和608可在实施例三提供的方法所包括的任意步骤之间执行，607和608的执行取决于关闭标识的发送时刻。其中，所述停止在所述第二低延迟转发模式下发送所述第二业务流的数据报文可采用方式一或方式二。方式一：关闭所述第二低延迟转发模式，且将在关闭所述第二低延迟转发模式后获得的所述第二业务流的数据报文送至低优先级的队列中。所述低优先级的队列和所述高优先级的队列是个相对的概念，所述低优先级的队列是相对于所述第二低延迟转发模式下的优先级队列而言的普通队列。方式二：关闭所述第二低延迟转发模式，且丢弃在关闭所述第二低延迟转发模式后获得的所述第二业务流的数据报文。所述关闭所述第二低延迟转发模式可以是获得所述关闭标识后。
本申请实施例三提供的方法中，PE1可对PE2上与所述第二业务流相关的门控进行灵活地控制，即发送用于控制所述第二低延迟转发模式的第二控制信息，无需通过管理平面为每个端口配置门控列表，降低管理的复杂性。
在图2所示的运营商网络为IP网络的情况下，PE1沿第一转发路径，通过P1，向PE2发送属于第一业务流的数据报文的方法，与上述PE1沿所述第二转发路径，向PE2发送属于所述第二业务流的数据报文的方法相似，在此不再赘述。所述第一转发路径为图2中运营商网络为IP网络时实线所表示的路径。
实施例四
图7为本申请实施例四提供的用于检测转发路径的传输时延的方法流程图。本申请实施例四是以第一转发路径为例，对用于检测转发路径的低延迟业务流的方法进行说明。实施例四提供的方法可在所述第一转发路径上的设备设置了第一低延迟转发模式之后，且在所述第一转发路径上的设备处理第一业务流之前执行。下面结合图2和图7，对本申请实施例四提供的用于检测转发路径的传输时延的方法进行说明。
701，PE1向P1发送第一检测报文。
举例说明，所述第一检测报文包括第一时延值和第二时延值。所述第一检测报文用于获得第一转发路径上转发设备产生的传输延迟。所述第一时延值用于表示所述第一转发路径上的转发设备允许的延迟的最大时长。所述第二时延值为PE1产生的延迟时长。PE1产生的延迟时长可以是通过实际测量计算获得PE1转发所述第一检测报文所产生的延迟值。
可选地，所述第一检测报文还包括PE1发送所述第一检测报文的时刻。这样，作为所述第一转发路径的出口的转发设备，比如PE2，可根据PE1发送所述第一检测报文的时刻和PE2接收到第二检测报文的时刻，获得所述第一转发路径的传输延迟。
其中，PE1发送所述第一检测报文的方法可与实施例二或实施例三中PE1发送第二数据报文的方法相同，在此不再赘述。
702，P1根据所述第一检测报文，获得第二检测报文。
举例说明，P1根据所述第一检测报文，获得第二检测报文包括：P1获得第三时延值，所述第三时延值为P1接收到所述第一检测报文到发送所述第一检测报文间的时长；P1根据所述第一检测报文和所 述第三时延值，获得所述第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。其中，P1发送所述第一检测报文的时刻即为P1发送所述第二检测报文的时刻，P1可以通过实际测量计算获得发送所述第一检测报文的时刻。
703，P1向PE2发送所述第二检测报文。
其中，P1发送所述第二检测报文的方法可与实施例二或实施例三中P1发送第三数据报文的方法相同，在此不再赘述。
704，PE2根据所述第二检测报文，获得第三检测报文。
举例说明，PE2根据所述第二检测报文，获得第三检测报文包括：PE2获得第四时延值，所述第四时延值为PE2接收到所述第二检测报文到发送所述第二检测报文间的时长；PE2根据所述第二检测报文和所述第四时延值，获得所述第三检测报文，所述第三检测报文包括所述第二检测报文和所述第四时延值。其中，PE2发送所述第二检测报文的时刻即为PE2发送所述第三检测报文的时刻，PE2可以预估出发送所述第二检测报文的时刻。
可选地，704之后，本申请实施例四提供的方法还包括：PE2上报所述第三检测报文。举例说明，PE2可向用于进行管理的设备发送所述第三检测报文，或者PE2向所述用于进行管理的设备发送所述第三检测报文中携带的信息和/或参数，以便所述用于进行管理的设备获得所述第一转发路径的传输时延。所述用于进行管理的设备可以是网管平台。
本申请实施例四提供的方法中，PE1沿第一转发路径发送第一检测报文，有助于通过所述第一转发路径上的设备对所述第一检测报文的处理，获得各个设备产生的延迟值，以定位无法满足第一延迟值的节点，有助于获得最佳的低延迟业务的转发效果。
实施例五
图8为本申请实施例五提供的第一转发设备的示意图。本申请实施例五提供的第一转发设备可以是图3、图4或图6中的PE1。本申请实施例五提供的第一转发设备可采用PE1所采用的方法。对于与实施例一至四任意一个实施例中相同的内容，在此不再赘述。下面结合图8，对本申请实施例五提供的第一转发设备进行说明。
本申请实施例五提供的第一转发设备包括处理单元801、获得单元802和第一发送单元803。
所述处理单元801用于确定接收的第一数据报文属于第一业务流后，获得与所述第一业务流对应的低延迟标识。
所述获得单元802用于在所述处理单元801确定所述第一数据报文属于所述第一业务流后，根据所述第一数据报文和所述低延迟标识，获得第二数据报文。所述第二数据报文包括所述第一数据报文和所述低延迟标识。所述低延迟转发模式是在动态控制下实现所述第一业务流的快速转发的模式。所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流。所述第二数据报文属于所述第一业务流。
所述第一发送单元803用于在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。若所述第一转发设备沿第一转发路径发送所述第二数据，则本实施例中的第二转发设备可以是图3或图4中的P1，P1是PE1在所述第一转发路径上的下一跳。若所述第一转发设备沿第二转发路径发送所述第二数据，则本实施例中的第二转发设备还可以是图3或图6中的PE2，PE2是PE1在所述第二转发路径上的下一跳。
可选地，所述获得单元802还用于获得控制信息。所述第一转发设备还包括：第二发送单元804。所述第二发送单元804用于向所述第二转发设备发送所述控制信息，所述控制信息用于控制所述低延迟转发模式的状态。
举例说明，所述控制信息的内容可参见实施例一中的相应内容。
可选地，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式，所述第一转发设备还包括第三发送单元805。所述第三发送单元805用于向所述第二转发设备发送关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式。
可选地，所述获得单元802还用于获得检测报文。所述第一转发设备还包括：第四发送单元806。所述第四发送单元806用于向所述第二转发设备发送所述检测报文，所述检测报文包括第一时延值和第二延迟值，所述检测报文用于获得转发路径上转发设备的传输延迟，所述第一时延值为所述转发路径上的转发设备允许的延迟的最大时长，所述第二延迟值为所述第一转发设备产生的延迟时长。举例说明，若所述第一转发设备为多条转发路径的入口节点，比如所述第一转发设备和所述第二转发设备的入口节点，所述第一转发设备可分别沿所述第一转发路径发送一个检测报文，沿所述第二转发路径发送另一个检测报文。
本申请实施例提供的第一转发设备可对第二转发设备上与所述第一业务流相关的门控进行灵活地控制，即发送用于控制低延迟转发模式的第一控制信息，无需通过管理平面为每个端口配置门控列表，降低管理的复杂性。
实施例六
图9为本申请实施例六提供的一种第二转发设备的示意图。本申请实施例六提供的第二转发设备可以是图3或图4的P1，P1是PE1在第一转发路径上的下一跳。所述第二转发设备还可以是图3或图4的PE2，PE2是PE1在第二转发路径上的下一跳。本申请实施例六提供的第二转发设备可采用P1或PE2所采用的方法，对于与实施例一至四任意一个实施例中相同的内容，在此不再赘述。下面结合图9，对本申请实施例六提供的第二转发设备进行说明。
所述第二转发设备包括：第一接收单元901和第一发送单元902。
所述第一接收单元901用于接收来自第一转发设备的第二数据报文。所述第二数据报文包括第一数据报文和低延迟标识。所述低延迟转发模式是在动态控制下实现所述第一业务流的快速转发的模式。所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流。所述第二数据报文属于所述第一业务流。
所述第一发送单元902用于根据所述低延迟标识，在低延迟转发模式下发送所述第二数据报文。若所述第二转发设备是中间转发设备，比如P1，则所述第一发送单元902可在所述低延迟转发模式下，向下一跳发送所述第二数据报文。若所述第二转发设备是作为网络出口的节点，比如PE2，则所述第一发送单元902可在所述低延迟转发模式下，向与其通信的CE发送所述第二数据报文包括的所述第一数据报文。
可选地，所述第二转发设备还包括：第二接收单元903和控制单元904。所述第二接收单元903用于接收来自所述第一转发设备的控制信息。所述控制信息用于控制所述低延迟转发模式的状态。所述控制单元904用于根据所述控制信息，控制所述低延迟转发模式的状态。
举例说明，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻。所述控制单元904具体用于根据所述低延迟转发模式的开始时刻和所述结束时刻，运行所述低延迟转发模式。
举例说明，所述控制信息包括所述低延迟转发模式的开始时刻和运行时长。所述控制单元904具体用于根据所述低延迟转发模式的开始时刻和所述运行时长，运行所述低延迟转发模式。
可选地，所述控制信息包括开启标识。所述开启标识用于标识开启所述低延迟转发模式。所述第二转发设备还包括：第三接收单元905。所述第三接收单元905用于接收所述第一转发设备发送的关闭标识。所述关闭标识用于标识关闭所述低延迟转发模式。所述控制单元904还用于根据所述关闭标识，控 制所述第一发送单元902停止在所述低延迟转发模式发送所述一业务流的数据报文。
可选地，所述第二转发设备还包括：第四接收单元906、第一获得单元907和第二获得单元908。所述第四接收单元906用于接收来自所述第一转发设备的第一检测报文，所述第一检测报文包括第一时延值和第二时延值，所述第一检测报文用于获得转发路径上的转发设备产生的传输延迟，所述第一时延值为所述转发路径上的转发设备可允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长。所述第一获得单元907用于获得第三时延值，所述第三时延值为所述第二转发设备接收到所述第一检测报文到发送所述第一检测报文间的时长。所述第二获得单元908用于根据所述第一检测报文和所述第三时延值，获得第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。
可选地，所述第二转发设备为中间转发设备，所述第一发送单元902具体用于向第三转发设备发送所述第二检测报文，所述第三转发设备为所述转发路径上沿第一方向所述第二转发设备的下一跳，所述第一方向为所述第一转发设备到所述转发路径上作为出口的转发设备的方向。
本申请实施例二中的第二转发设备可在接收第一数据报文后，根据配置的低延迟转发模式，在低延迟转发模式下发送所述所述第二数据报文，无需存储管理平面下发的门控列表，实现较为灵活地控制，还有助于降低管理的复杂性。
实施例七
图10为本申请实施例七提供的一种第一转发设备的示意图。图10所示的第一转发设备和图8所示的第一转发设备可以是同一台设备。图10所示的第一转发设备可以是上述实施例一至实施例四任意一个实施例中的PE1。本申请实施例七提供的第一转发设备包括：处理器1001、存储器1002和通信接口1003。所述处理器1001、所述存储器1002和所述通信接口1003通过通信总线1004连接。所述存储器1002用于存储程序。可选地，存储器1002还可用于存储控制信息和/或实施例二中的特征信息表。
所述处理器1001根据从所述存储器1002中读取的程序所包括的可执行指令，执行如下操作：
通过所述通信接口1003，接收第一数据报文；
确定所述第一数据报文属于第一业务流；
根据所述第一数据报文和低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述第二数据报文属于所述第一业务流；
通过所述通信接口1003，在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。
可选地，所述处理器1001还可根据所述可执行指令，进一步通过所述通信接口1003，向所述第二转发设备发送控制信息。所述控制信息用于控制所述低延迟转发模式的状态。
举例说明，所述控制信息携带于RSVP消息或G-ACH通道消息。所述RSVP消息和G-ACH通道消息的具体描述可参见实施例一中的相应内容，在此不再赘述。
举例说明，所述控制信息的内容可参见实施例一中的相应内容，在此不再赘述。
可选地，所述处理器1001还可根据所述可执行指令，在向所述第二转发设备发送所述第二数据报文之前，通过所述通信接口1003，向所述第二转发设备发送检测报文。所述检测报文包括第一时延值和第二时延值。所述第一时延值和所述第二时延值的内容与上述实施例相同，在此不再赘述。
在一种方式中，所述通信接口1003包括至少两个逻辑接口、至少一个队列和至少一个门，图10仅以所述通信接口1003包括第一逻辑接口10030、队列10031、门10032和第二逻辑接口10033为例进行说明。所述队列10031、所述门10032和所述第二逻辑接口10033用于处理所述第一业务流的数据报文。所述处理器1001可通过第一逻辑接口10030接收所述第一数据报文。本实施例中的第一逻辑接口10030可以是实施例二中的第一端口或第二端口。所述处理器1001可将获得的第二数据报文输出至所 述队列10031。所述处理器1001根据所述控制信息，控制所述门10032的开启，即切换至所述低延迟转发模式下。所述处理器1001控制所述队列10031中第一业务流的数据报文，比如所述第二数据报文，优先输出至所述门10032。所述门10032处于开启状态，所述第二数据报文被送至第二逻辑接口10033，并通过所述第二逻辑接口10033输出至相应的物理线路。
可选地，所述通信接口1003还包括能够与所述第二逻辑接口10033通信的队列10032。所述队列10032为按通常的转发模式进行处理的待转发数据报文形成的队列。所述第二转发设备停止采用所述低延迟转发模式后，可通过所述队列10032和所述第二逻辑接口10033，采用通常的队列转发方式进行数据报文的发送。其中，所述队列10032的优先级低于所述队列10031的优先级。
其中，所述第二逻辑接口10033和所述第一逻辑接口10030在物理上可以是同一个物理接口，该物理接口能够实现收发功能。所述第二逻辑接口10033和所述第一逻辑接口10030在物理上也可通过不同的物理接口来实现，本申请实施例对其具体实现方式不进行限定。
在一种方式中，所述处理器1001可根据其内部的时钟以及所述控制信息，控制所述门10032的开启。在另一种方式中，所述第一转发设备还包括时钟(timer)1005。时钟1005可用于对所述处理器1001和所述门10032的时钟进行同步。所述处理器1001可根据时钟1005及所述控制信息，控制所述门10032的开启。
实施例八
图11为本申请实施例八提供的一种第二转发设备的示意图。图11所示的第二转发设备和图9所示的第二转发设备可以是同一台设备。图11所示的第二转发设备可以上述实施例一至实施例四任意一个实施例中的P1或PE2。本申请实施例八提供的第二转发设备包括：处理器1101、存储器1102和通信接口1103。所述处理器1101、所述存储器1102和所述通信接口1103通过通信总线1104连接。所述存储器1102用于存储程序。可选地，存储器1102还可用于存储控制信息。
所述处理器1101根据从所述存储器1102中读取的程序所包括的可执行指令，执行如下操作：
通过所述通信接口1103，接收来自第一转发设备的第二数据报文，所述第二数据报文包括第一数据报文和低延迟标识，所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述第二数据报文属于所述第一业务流；
根据所述低延迟标识，通过所述通信接口1103，在低延迟转发模式下发送所述第二数据报文
可选地，所述处理器1101还可根据所述可执行指令，进一步通过所述通信接口1103，接收来自所述第一转发设备的控制信息。所述控制信息用于控制所述低延迟转发模式的状态。所述处理器1101进一步根据所述控制信息，控制所述低延迟转发模式的状态。
举例说明，所述控制信息可携带于RSVP消息或G-ACH通道消息，具体可参见实施例一中的相应内容。
可选地，所述控制信息的内容可参见实施例一中的相应内容，在此不再赘述。所述处理器1101根据所述控制信息动态控制所述低延迟转发模式的方法可参见实施例一或实施例二中的相应内容，在此不再赘述。
可选地，所述处理器1101还可根据所述可执行指令，进一步通过所述通信接口1103，接收来自所述第一转发设备的第一检测报文。所述第一检测报文包括第一时延值和第二时延值。所述处理器1101进一步获得第三时延值。所述第一时延值、所述第二时延值和所述第三时延值与上述实施例中的相同，在此不再赘述。所述处理器1101进一步根据所述第一检测报文和所述第三时延值，获得第二检测报文。所述第二检测报文包括所述第一检测报文和所述第三时延值。
可选地，若所述第二转发设备为中间转发设备，比如实施例一、实施例二或实施例四中的P1，则 所述处理器1101还可根据所述可执行指令，进一步通过所述通信接口1103，向第三转发设备发送所述第二检测报文。所述第三转发设备为所述转发路径上所述第二转发设备的下一跳。
在一种方式中，所述通信接口1103包括至少两个逻辑接口、至少一个队列和至少一个门，图11仅以所述通信接口1103包括第一逻辑接口11030、队列11031、门11032和第二逻辑接口10033为例进行说明。所述队列11031、所述门11032和所述第二逻辑接口10033用于处理所述第一业务流的数据报文。所述处理器1101可通过第一逻辑接口11030接收所述第二数据报文。所述处理器1101可将所述第二数据报文输出至所述队列11031。所述处理器1101根据所述控制信息，控制所述门11032的开启，即切换至所述低延迟转发模式下。所述处理器1101控制所述队列11031中第一业务流的数据报文，比如所述第二数据报文，优先输出至所述门11032。所述门11032处于开启状态，所述第二数据报文被送至所述第二逻辑接口11033，通过所述第二逻辑接口11033输出至相应的物理线路。
可选地，所述通信接口1103还包括能够与所述第二逻辑接口11033通信的队列11032。所述队列11032为按通常的转发模式进行处理的待转发数据报文形成的队列。所述第二转发设备停止采用所述低延迟转发模式后，可通过所述队列11032和所述第二逻辑接口11033，采用通常的队列转发方式进行数据报文的发送。其中，所述队列11032的优先级低于所述队列11031的优先级。
其中，所述第二逻辑接口11033和所述第一逻辑接口11030在物理上可以是同一个物理接口，该物理接口能够实现收发功能。所述第二逻辑接口11033和所述第一逻辑接口11030在物理上也可通过不同的物理接口来实现，本申请实施例对其具体实现方式不进行限定。
在一种方式中，所述处理器1101可根据其内部的时钟以及所述控制信息，控制所述门11032的开启。在另一种方式中，所述第一转发设备还包括时钟(timer)1105。所述处理器1101可根据时钟1105及所述控制信息，控制所述门11032的开启。时钟1105可用于对所述处理器1101和所述门11032的时钟进行同步。
本申请实施例九还提供了一种用于处理低延迟业务流的系统。本申请实施例九提供的系统可包括实施例五提供的第一转发设备和实施例六提供的第二转发设备。或者，本申请实施例提供的系统可包括实施例七提供的第一转发设备和实施例八提供的第二转发设备。在此不再对每个转发设备的功能和结构进行赘述，具体可参见相应实施例中的描述。
实施例十
图12为本申请实施例十提供的用于建立转发路径的方法流程图。实施例十提供的方法可用于建立图2所示的网络场景中的第一转发路径或第二转发路径。实施例十提供的方法可以由图2所示的运营商网络中的任意一个节点执行。实施例十提供的方法中的第一节点为图2所示的运营商网络中的任意一个节点。例如所述第一节点可以为图2中的PE1、P1或PE2。本申请实施例十提供的方法可从图13所示的网络场景中，确定出图2所示的第一转发路径和第二转发路径。下面结合图13和图2所示的网络场景，对本申请实施例十提供的方法进行说明。
1201，第一节点获得运营商网络中的拓扑信息。
其中，所述拓扑信息为入口节点与出口节点之间的每条路径上的节点的拓扑信息。所述拓扑信息包括相邻两个节点之间的物理链路时延和每个节点的节点驻留时长。本申请实施例中的入口节点为作为网络入口的节点。本申请实施例中的出口节点为作为网络出口的节点。
1202，所述第一节点根据所述拓扑信息，获得所述入口节点和所述出口节点之间每条路径的传输时延。
1203，所述第一节点根据所述每条路径的传输时延，确定所述入口节点和所述出口节点之间的目标传输路径。
如图2和图13所示，入口节点为PE1，出口节点为PE2。PE1和PE2之间的目标传输路径可以是上述实施例中提及的第一转发路径和/或第二转发路径。所述第一转发路径的传输时延为所述第一转发路径上相邻两个节点之间的物理链路时延和所述第一转发路径上每个节点的节点驻留时长之和。如图2所示，所述第一转发路径的传输时延为PE1和P1之间的物理链路时延、P1和PE2之间的物理链路时延、PE1的节点驻留时长、P1的节点驻留时长和PE2的节点驻留时长之和。所述第二转发路径的传输时延为所述第二转发路径上相邻两个节点之间的物理链路时延和所述第二转发路径上每个节点的节点驻留时长之和。如图2所示，所述第二转发路径的传输时延为PE1和PE2之间的物理链路时延、PE1的节点驻留时长和PE2的节点驻留时长之和。
对于1201来说，以所述运营商网络中的任意一个节点为例，所述任意一个节点可以测量其与邻居节点之间的物理链路时延，还可以测量所述任意一个的节点驻留时长。所述任意一个节点可在测量得到其物理链路时延和其节点驻留时长后，以拓扑信息的形式，将所述任意一个节点获得的物理链路时延以及所述任意一个节点的节点驻留时长发送给所述运营商网络中的其他节点。本申请实施例中提及的节点驻留时长是该节点产生的延迟值，或者为该节点产生的延迟时长。
对于1201来说，如图2所示，所述第一节点可接收P1发送的拓扑信息。所述P1发送的拓扑信息包括P1与PE1间的物理链路时延、P1与PE2间的物理链路时延和P1的节点驻留时长。所述第一节点还可接收PE2发送的拓扑信息。PE2发送的拓扑信息包括P1与PE2间的物理链路时延、PE2与PE1件的物理链路时延和PE2的节点驻留时长。所述第一节点可接收PE1发送的拓扑信息。所述PE1发送的拓扑信息包括P1与PE1间的物理链路时延、PE1与PE2间的物理链路时延和PE1的节点驻留时长。所述第一节点可以是图2所示的运营商网络中除PE1、P1和PE2之外的节点，还可以是PE1、P1和PE2中的任意一个。所述第一节点可根据获得的拓扑信息，确定出入口节点和出口节点之间的目标传输路径。所述目标传输路径为满足时延要求的用于传输低延迟业务流的路径，比如图2所示的第一转发路径或第二转发路径。
1201中，以所述第一节点获得所述第一拓扑信息为例进行说明。所述第一节点可以通过接收来自所述第一邻居节点的时延测量报文进行所述第一物理链路时延的测量。例如：所述第一节点接收所述第一邻居节点发送的时延测量报文。所述时延测量报文包括所述第一邻居节点发送所述时延测量报文的发送时间戳。所述第一节点可以根据接收所述延时测量报文的接收时间戳与所述发送时间戳，获得所述第一物理链路时延。所述延时测量报文是由所述第一邻居节点不经过其他节点转发而直接发送至所述第一节点。如图13所示，若第一节点为图13中的节点4，第一邻居节点为图13中的节点6，则所述节点6向所述节点4发送时延测量报文，所述时延测量报文中包括所述节点6发送所述时延测量报文的发送时间戳t1。所述节点4接收到所述延迟测量报文后，获得接收时间戳t2。所述节点4根据接收时间戳t2与所述时延测量报文包括的发送时间戳t1，可以获得所述节点6和所述节点4之间的物理链路时延。图13中任意两个相邻的节点获得其物理链路时延的方法与上述节点6与节点4间物理链路时延的获得方法相同，在此不再逐一举例说明。图13中的节点1可以是图2中的P1。
可选地，对于任意两个相邻节点之间的物理链路时延，任意两个相邻节点之间的物理链路时延的数值不会因为方向不同而不同。如果任意两个相邻节点之间的物理链路时延的数值有可能因为方向不同而存在差异，则可对任意两个相邻节点间不同方向的物理链路时延分别进行检测，可将两个方向的检测结果的平均值作为所述任意两个相邻节点间的物理链路时延。
1201中，对于任意两个相邻节点之间的物理链路时延的测量，可以通过发送多个延迟测量报文，进行多次的测量。对所述多次测量结果进行统计，例如取平均值、计算期望值、取最大值或取最小值等，获得所述两个相邻节点之间的物理链路时延。
本申请实施例中，报文在节点内的节点驻留时长与节点的实际负载密切相关。在一个空载的节点上，报文的节点驻留时长非常小。在一个接近满载的节点上，报文的节点驻留时长比较大。因此，为了更精确地反映报文的节点驻留时长，1201中，可以通过查询节点负载与驻留时长之间的映射表，获取节点驻留时长。另外，节点驻留时长还可以由网络拓扑结构、链路介质、长度以及节点的设备决定。具体地，可以在各个节点中预先建立所述节点的节点负载与驻留时长之间的映射表。例如，对于任意一个节点，所述节点内的负载与驻留时长的映射表可以为表1。
节点负载(百分比) 节点驻留时长(毫秒)
0％ 0ms
10％ 0.01ms
20％ 0.05ms
… …
50％ 0.5ms
表1
在所述节点的负载为20％的情况下，所述节点的节点驻留时长即为0.05ms。任意一个节点可以获得其负载情况，根据其负载情况查询映射表，获得节点驻留时长。其中，所述任意一个节点的负载可以为任意时刻的负载，或者为某一段时间内的负载的平均值，或者为取某一时刻的负载。
在本申请实施例中，任意一个节点利用开放式最短路径优先(Open Shortest Path First，OSPF)或IS-IS协议传输所述任意一个节点获得的拓扑信息。所述任意一个节点获得拓扑信息可以是其自身检测获得的拓扑信息，还可以来自其他节点的拓扑信息。举例说明，可对OSPF或IS-IS协议进行路由能力扩展，使其携带所述任意一个节点获得的拓扑信息，在此不再对可能的结构进行赘述。
1203中，所述第一节点可以根据所述入口节点和所述出口之间每条路径的传输时延，确定满足时延要求的传输路径为所述目标传输路径。可选地，所述目标传输路径可以为一条路径，也可以为满足所述时延要求的多条路径。所述时延要求可以为低延迟业务流所要求的传输时延小于或等于预设值。如图13所示，PE1为入口节点，PE2为出口节点，PE1与PE2之间存在多条传输路径。所述第一节点可计算图13中每条传输路径的传输时延。图2中的第一转发路径(实线)为图13中多条传输路径中时延较小且满足所述第一业务流的时延要求的传输路径。图2中的第二转发路径(虚线)为图13中多条传输路径中时延较小且满足所述第二业务流的时延要求的传输路径。所述第一节点可确定图2中的第一转发路径(实线)和第二转发路径(虚线)为目标传输路径。
本申请实施例十中的节点可以是图2所示的运营商网络中的设备。
上述通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器。结合本申请实施例所公开的方法的步骤，可以直接体现为处理器中的硬件及软件模块组合执行完成。当使用软件实现时，可以将实现上述功能的代码存储在计算机可读介质中。计算机可读介质包括计算机存储介质。存储介质可以是计算机能够存取的任何可用介质。以此为例但不限于：计算机可读介质可以是随机存取存储器(英文全称为random-access memory，英文缩写为RAM)、只读存储器(英文全称为read-only memory，英文缩写为ROM)、电可擦可编程只读存储器(英文全称为electrically erasable programmable read-only memory，英文缩写为EEPROM)、只读光盘(英文全称为compact disk read-only memory，英文缩写为CD-ROM)或其他光盘存储、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质。计算机可读介质可以是压缩碟(英文全称为compact disk，英文缩写为CD)、激光碟、数字视频光碟(英文全称为digital video disk，英文缩写为DVD)、软盘或者蓝光碟。
最后应说明的是：以上实施例仅用以示例性说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请及本申请带来的有益效果进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请权利要求的范围。

Claims (24)

  1. 一种用于处理低延迟业务流的方法，其特征在于，所述方法包括：
    第一转发设备确定接收到的第一数据报文属于第一业务流后，获得与所述第一业务流对应的低延迟标识，所述第一转发设备为作为网络入口的设备；
    所述第一转发设备根据所述第一数据报文和所述低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式；
    所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。
  2. 根据权利要求1所述的方法，其特征在于，所述第一转发设备在所述低延迟转发模式下，向所述第二转发设备发送所述第二数据报文之前，所述方法还包括：
    所述第一转发设备向所述第二转发设备发送控制信息，所述控制信息用于控制所述低延迟转发模式的状态。
  3. 根据权利要求2所述的方法，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，或者
    所述控制信息包括所述低延迟转发模式的开始时刻和运行时长。
  4. 根据权利要求2所述的方法，其特征在于，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式；在所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之后，所述方法还包括：
    所述第一转发设备向所述第二转发设备发送关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式。
  5. 根据权利要求1至4任一所述的方法，其特征在于，在所述第一转发设备在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文之前，所述方法还包括：
    所述第一转发设备向所述第二转发设备发送检测报文，所述检测报文包括第一时延值和第二时延值，所述第一时延值为转发路径上的转发设备允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长。
  6. 一种用于处理低延迟业务流的方法，其特征在于，所述方法包括：
    第二转发设备接收来自第一转发设备的第二数据报文，所述第二数据报文包括低延迟标识，所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式，所述第二数据报文属于所述第一业务流；
    所述第二转发设备根据所述低延迟标识，在所述低延迟转发模式下处理所述第二数据报文。
  7. 根据权利要求6所述的方法，其特征在于，所述第二转发设备接收来自第一转发设备的第二数据报文之前，所述方法还包括：
    所述第二转发设备接收来自所述第一转发设备的控制信息，所述控制信息用于控制所述低延迟转发模式的状态；
    所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态。
  8. 根据权利要求7所述的方法，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态包括：
    所述第二转发设备根据所述低延迟转发模式的开始时刻和所述结束时刻，运行所述低延迟转发模 式。
  9. 根据权利要求7所述的方法，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和运行时长，所述第二转发设备根据所述控制信息，动态控制所述低延迟转发模式的状态包括：
    所述第二转发设备根据所述低延迟转发模式的开始时刻和所述运行时长，运行所述低延迟转发模式。
  10. 根据权利要求7所述的方法，其特征在于，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式；所述第二转发设备根据所述低延迟标识，在低延迟转发模式下发送所述第二数据报文之后，所述方法还包括：
    所述第二转发设备接收所述第一转发设备发送的关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式；
    所述第二转发设备根据所述关闭标识，停止在所述低延迟转发模式发送所述第一业务流的数据报文。
  11. 根据权利要求6至10任一所述的方法，其特征在于，所述方法还包括：
    所述第二转发设备接收来自所述第一转发设备的第一检测报文，所述第一检测报文包括第一时延值和第二时延值，所述第一时延值为转发路径上的转发设备可允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长；
    所述第二转发设备获得第三时延值，所述第三时延值为所述第二转发设备接收到所述第一检测报文到发送所述第一检测报文间的时长；
    所述第二转发设备根据所述第一检测报文和所述第三时延值，获得第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。
  12. 根据权利要求11所述的方法，其特征在于，所述第二转发设备为中间转发设备，所述方法还包括：
    所述第二转发设备向第三转发设备发送所述第二检测报文，所述第三转发设备为所述转发路径上沿第一方向所述第二转发设备的下一跳，所述第一方向为所述第一转发设备到作为网络出口的设备的方向。
  13. 一种第一转发设备，其特征在于，所述第一转发设备包括：
    处理单元，用于确定接收到的第一数据报文属于第一业务流后,获得与所述第一业务流对应的低延迟标识，所述第一转发设备为作为网络入口的设备；
    获得单元，用于根据所述第一数据报文和所述低延迟标识，获得第二数据报文，所述第二数据报文包括所述第一数据报文和所述低延迟标识，所述低延迟标识用于指示接收到所述第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式；
    第一发送单元，用于在所述低延迟转发模式下，向第二转发设备发送所述第二数据报文。
  14. 根据权利要求13所述的第一转发设备，其特征在于，所述第一转发设备还包括：
    第二发送单元，用于向所述第二转发设备发送控制信息，所述控制信息用于控制所述低延迟转发模式的状态。
  15. 根据权利要求14所述的第一转发设备，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，或者
    所述控制信息包括所述低延迟转发模式的开始时刻和运行时长。
  16. 根据权利要求14所述的第一转发设备，其特征在于，所述控制信息包括开启标识，所述开启 标识用于标识开启所述低延迟转发模式，所述第一转发设备还包括：
    第三发送单元，用于向所述第二转发设备发送关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式。
  17. 根据权利要求13至16任一所述的第一转发设备，其特征在于，所述第一转发设备还包括：
    第四发送单元，用于向所述第二转发设备发送检测报文，所述检测报文包括第一时延值和第二时延值，所述第一时延值为转发路径上的转发设备允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长。
  18. 一种第二转发设备，其特征在于，所述第二转发设备包括：
    第一接收单元，用于接收来自第一转发设备的第二数据报文，所述第二数据报文包括低延迟标识，所述低延迟标识用于指示接收到第一业务流的转发设备在低延迟转发模式下转发所述第一业务流，所述低延迟转发模式是动态控制下实现所述第一业务流的快速转发的模式，所述第二数据报文属于所述第一业务流；
    第一发送单元，用于根据所述低延迟标识，在所述低延迟转发模式下发送所述第二数据报文。
  19. 根据权利要求18所述的第二转发设备，其特征在于，所述第二转发设备还包括：
    第二接收单元，用于接收来自所述第一转发设备的控制信息，所述控制信息用于控制所述低延迟转发模式的状态；
    控制单元，用于根据所述控制信息，动态控制所述低延迟转发模式的状态。
  20. 根据权利要求19所述的第二转发设备，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和结束时刻，
    所述控制单元具体用于根据所述低延迟转发模式的开始时刻和所述结束时刻，运行所述低延迟转发模式。
  21. 根据权利要求19所述的第二转发设备，其特征在于，所述控制信息包括所述低延迟转发模式的开始时刻和运行时长，
    所述控制单元具体用于根据所述低延迟转发模式的开始时刻和所述运行时长，运行所述低延迟转发模式。
  22. 根据权利要求19所述的第二转发设备，其特征在于，所述控制信息包括开启标识，所述开启标识用于标识开启所述低延迟转发模式，所述第二转发设备还包括：
    第三接收单元，用于接收所述第一转发设备发送的关闭标识，所述关闭标识用于标识关闭所述低延迟转发模式；
    所述控制单元还用于根据所述关闭标识，停止在所述低延迟转发模式发送所述第一业务流的数据报文。
  23. 根据权利要求18至22任一所述的第二转发设备，其特征在于，所述第二转发设备还包括：
    第四接收单元，用于接收来自所述第一转发设备的第一检测报文，所述第一检测报文包括第一时延值和第二时延值，所述第一时延值为转发路径上的转发设备可允许的延迟的最大时长，所述第二时延值为所述第一转发设备产生的延迟时长；
    第一获得单元，用于获得第三时延值，所述第三时延值为所述第二转发设备接收到所述第一检测报文到发送所述第一检测报文间的时长；
    第二获得单元，用于根据所述第一检测报文和所述第三时延值，获得第二检测报文，所述第二检测报文包括所述第一检测报文和所述第三时延值。
  24. 根据权利要求23所述的第二转发设备，其特征在于，所述第二转发设备为中间转发设备，所述第一发送单元具体用于向第三转发设备发送所述第二检测报文，所述第三转发设备为所述转发路径上沿第一方向所述第二转发设备的下一跳，所述第一方向为第一转发设备到作为网络出口的设备的方向。
PCT/CN2017/088695 2016-08-27 2017-06-16 一种用于处理低延迟业务流的方法和装置 WO2018040673A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP17844979.9A EP3468123B1 (en) 2016-08-27 2017-06-16 Method and device for use in processing low-latency traffic
EP21206178.2A EP4030732A1 (en) 2016-08-27 2017-06-16 Method and apparatus for processing low-latency service flow
US16/285,913 US10972398B2 (en) 2016-08-27 2019-02-26 Method and apparatus for processing low-latency service flow
US17/222,201 US11616729B2 (en) 2016-08-27 2021-04-05 Method and apparatus for processing low-latency service flow

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610743336.6 2016-08-27
CN201610743336.6A CN107786465B (zh) 2016-08-27 2016-08-27 一种用于处理低延迟业务流的方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/285,913 Continuation US10972398B2 (en) 2016-08-27 2019-02-26 Method and apparatus for processing low-latency service flow

Publications (1)

Publication Number Publication Date
WO2018040673A1 true WO2018040673A1 (zh) 2018-03-08

Family

ID=61299933

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/088695 WO2018040673A1 (zh) 2016-08-27 2017-06-16 一种用于处理低延迟业务流的方法和装置

Country Status (4)

Country Link
US (2) US10972398B2 (zh)
EP (2) EP3468123B1 (zh)
CN (2) CN107786465B (zh)
WO (1) WO2018040673A1 (zh)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10594612B2 (en) * 2015-09-04 2020-03-17 Nokia Technologies Oy Threshold for reduced latency mechanisms
CN113518031A (zh) * 2017-06-16 2021-10-19 华为技术有限公司 一种业务转发的方法及网络设备
EP3468118B1 (en) * 2017-10-04 2023-10-04 TTTech Computertechnik AG Method and device to configure real-time networks
CN115714742A (zh) * 2018-05-11 2023-02-24 华为技术有限公司 一种报文发送的方法、网络节点和系统
US11647445B2 (en) * 2018-07-23 2023-05-09 Lg Electronics Inc. V2X communication device and geo-networking transmission method
CN110876159B (zh) * 2018-08-30 2022-02-11 华为技术有限公司 一种提高时延确定性的方法及装置
US11223567B2 (en) * 2019-01-18 2022-01-11 Cisco Technology, Inc. Transmission control protocol session mobility
CN111526583B (zh) * 2019-02-01 2023-03-28 大唐移动通信设备有限公司 资源配置方法、获取方法、网络设备及终端
CN110233765B (zh) * 2019-06-23 2022-08-12 何梓菁 一种低时延网络切片方法和设备
CN110995595B (zh) * 2019-12-16 2022-03-11 新华三大数据技术有限公司 报文发送方法、装置、存储介质和节点设备
CN111049910A (zh) * 2019-12-16 2020-04-21 瑞斯康达科技发展股份有限公司 一种处理报文的方法、装置、设备及介质
US11362959B2 (en) * 2020-03-18 2022-06-14 Futurewei Technologies, Inc. Latency based forwarding of packets with destination policies
CN111510344B (zh) * 2020-04-02 2021-04-02 中国信息通信研究院 一种节点的转发时延确定方法和装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060239188A1 (en) * 2005-04-20 2006-10-26 Walter Weiss Providing a quality of service for various classes of service for transfer of electronic data packets
CN101488914A (zh) * 2009-01-06 2009-07-22 杭州华三通信技术有限公司 服务质量QoS的实现方法和服务边缘提供商PE设备
CN101645849A (zh) * 2009-09-09 2010-02-10 中兴通讯股份有限公司 一种在过渡环境中的QoS实现方法和PE路由器
CN104579962A (zh) * 2015-01-23 2015-04-29 盛科网络（苏州）有限公司 一种区分不同报文的QoS策略的方法及装置
CN105847176A (zh) * 2016-05-20 2016-08-10 锐捷网络股份有限公司 一种传输业务流的方法和网络设备

Family Cites Families (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0712220A1 (en) * 1994-11-08 1996-05-15 International Business Machines Corporation Hop-by-hop flow control in an ATM network
KR100333250B1 (ko) * 1998-10-05 2002-05-17 가나이 쓰토무 패킷 중계 장치
US6618357B1 (en) * 1998-11-05 2003-09-09 International Business Machines Corporation Queue management for networks employing pause time based flow control
US6643259B1 (en) * 1999-11-12 2003-11-04 3Com Corporation Method for optimizing data transfer in a data network
US6674717B1 (en) * 2000-03-30 2004-01-06 Network Physics, Inc. Method for reducing packet loss and increasing internet flow by feedback control
US7027396B1 (en) * 2002-02-13 2006-04-11 At&T Corp. Traffic matrix computation for a backbone network supporting virtual private networks
KR100440062B1 (ko) * 2002-05-20 2004-07-14 엘지전자 주식회사 비지피 엠피엘에스 브이피엔 망의 브이피엔 씨오에스 제공방법
GB0226249D0 (en) * 2002-11-11 2002-12-18 Clearspeed Technology Ltd Traffic handling system
US7626986B1 (en) * 2002-11-18 2009-12-01 At&T Corp. Method for operating a router having multiple processing paths
WO2004066120A2 (en) * 2003-01-15 2004-08-05 Ciena Coporation A method and apparatus for transporting packet data over an optical network
US7564792B2 (en) * 2003-11-05 2009-07-21 Juniper Networks, Inc. Transparent optimization for transmission control protocol flow control
US7953000B2 (en) * 2004-09-10 2011-05-31 Cisco Technology, Inc. Mechanism to improve preemption behavior of resource reservations
US7907519B2 (en) * 2005-01-28 2011-03-15 British Telecommunications Plc Packet forwarding
US7684391B2 (en) * 2006-01-13 2010-03-23 Comcast Cable Holdings, Llc Communications system for delivering multimedia internet protocol packets across network boundaries
EP2100390B1 (en) * 2006-10-22 2011-04-06 ViaSat, Inc. Cyclical obstruction communication system
US7697519B2 (en) * 2006-10-31 2010-04-13 Hewlett-Packard Development Company, L.P. Packet processing
JP4753909B2 (ja) * 2007-04-12 2011-08-24 アラクサラネットワークス株式会社 トラヒックシェーピング回路、端末装置及びネットワークノード
US9215620B2 (en) * 2008-05-05 2015-12-15 Cisco Technology, Inc. Distributed bi-directional flow control in wireless mesh networks
US8031607B2 (en) * 2009-01-29 2011-10-04 Alcatel Lucent Implementation of internet protocol header compression with traffic management quality of service
WO2010115096A2 (en) * 2009-04-02 2010-10-07 University Of Florida Research Foundation Inc. System, method, and media for network traffic measurement on high-speed routers
US8611327B2 (en) * 2010-02-22 2013-12-17 Broadcom Corporation Method and apparatus for policing a QoS flow in a MoCA 2.0 network
CN102238073A (zh) * 2010-04-30 2011-11-09 华为技术有限公司 网络业务流管理的方法和设备
CN102378382B (zh) * 2010-08-10 2015-05-27 华为技术有限公司 一种数据流的调度方法、设备和系统
CN102447980B (zh) * 2010-09-30 2015-01-28 中兴通讯股份有限公司 一种路由控制方法、系统及路由计算装置
US9124524B2 (en) * 2011-06-29 2015-09-01 Broadcom Corporation System and method for priority based flow control between nodes
US8638667B2 (en) * 2011-07-05 2014-01-28 Cisco Technology, Inc. Transmission priority paths in mesh networks
US8929220B2 (en) * 2012-08-24 2015-01-06 Advanced Micro Devices, Inc. Processing system using virtual network interface controller addressing as flow control metadata
US9509549B2 (en) * 2013-03-15 2016-11-29 Cisco Technology, Inc. Extending routing rules from external services
US10193801B2 (en) * 2013-11-25 2019-01-29 Juniper Networks, Inc. Automatic traffic mapping for multi-protocol label switching networks
EP3254462A1 (en) * 2015-02-05 2017-12-13 Cisco Technology, Inc. Method and system for processing hevc coded video in broadcast and streaming applications
EP3073680B1 (en) * 2015-03-23 2017-12-06 Alcatel Lucent Methods, queueing system, network element and network system for queueing and processing of packets
US20160295579A1 (en) * 2015-03-31 2016-10-06 Alcatel-Lucent Canada Inc. Systems and methods for enhancing real-time quality of experience in universal mobile telecommunications systems and long term evolution communications networks
US20160308786A1 (en) * 2015-04-17 2016-10-20 Futurewei Technologies, Inc. Temporal Tunnel Services
US10425348B2 (en) * 2015-07-22 2019-09-24 The Regents Of The University Of Colorado Stateless network functions
EP3440810B1 (en) * 2016-04-04 2020-12-30 Telefonaktiebolaget LM Ericsson (publ) Quality of service (qos) support for tactile traffic
AU2017300019B2 (en) * 2016-07-19 2022-04-14 Schneider Electric Industries Sas Time-sensitive software defined networking
US20180026933A1 (en) * 2016-07-22 2018-01-25 Cisco Technology, Inc. Service aware label address resolution protocol switched path instantiation
US10291517B1 (en) * 2016-12-16 2019-05-14 Juniper Networks, Inc. Generating a dummy VLAN tag for indicating quality of service classification information in a distributed routing system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060239188A1 (en) * 2005-04-20 2006-10-26 Walter Weiss Providing a quality of service for various classes of service for transfer of electronic data packets
CN101488914A (zh) * 2009-01-06 2009-07-22 杭州华三通信技术有限公司 服务质量QoS的实现方法和服务边缘提供商PE设备
CN101645849A (zh) * 2009-09-09 2010-02-10 中兴通讯股份有限公司 一种在过渡环境中的QoS实现方法和PE路由器
CN104579962A (zh) * 2015-01-23 2015-04-29 盛科网络（苏州）有限公司 一种区分不同报文的QoS策略的方法及装置
CN105847176A (zh) * 2016-05-20 2016-08-10 锐捷网络股份有限公司 一种传输业务流的方法和网络设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3468123A4 *

Also Published As

Publication number Publication date
US10972398B2 (en) 2021-04-06
EP3468123A1 (en) 2019-04-10
US20190199642A1 (en) 2019-06-27
US11616729B2 (en) 2023-03-28
CN113472673A (zh) 2021-10-01
CN107786465A (zh) 2018-03-09
CN107786465B (zh) 2021-06-04
EP3468123B1 (en) 2021-12-01
EP3468123A4 (en) 2019-06-05
US20210226897A1 (en) 2021-07-22
EP4030732A1 (en) 2022-07-20

Similar Documents

Publication Publication Date Title
WO2018040673A1 (zh) 一种用于处理低延迟业务流的方法和装置
US11848757B2 (en) In-situ passive performance measurement in a network environment
US10855574B2 (en) Method and network device for computing forwarding path
CN112262553B (zh) 在软件定义网络交换机的分组处理流水线中跟踪分组的装置和方法
WO2021170092A1 (zh) 报文处理方法、装置、网络设备及存储介质
KR102071439B1 (ko) 타깃 전송 경로를 획득하는 방법 및 네트워크 노드
US20230006906A1 (en) In-situ flow detection method and apparatus
US10965594B2 (en) Pseudo wire load sharing method and device
US12096272B2 (en) Transporting MTNC-ID over SRV6-header for 5G transport
EP3586482B1 (en) Mechanism to detect data plane loops in an openflow network
WO2014194711A1 (zh) 报文处理方法、设备标签处理方法及设备
WO2017000802A1 (zh) 一种业务故障定位方法及装置
WO2011160517A1 (zh) 一种多协议标签交换业务隧道切换的方法及系统
WO2012142753A1 (zh) Ip承载网性能监控的方法及设备
Sameh et al. Dealing with quality of service in hybrid wired-wireless networks
WO2020114083A1 (zh) 一种ioam信息的处理方法和装置
CN114598636A (zh) 流量调度方法、设备及系统
WO2012159351A1 (zh) 一种标签分配方法、装置和系统
WO2024153327A1 (en) Improved intent requests and proposals using proposal times and accuracy levels
Stein et al. Transport Network Aspects
US20140010238A1 (en) Method of communication between two items of termination equipment

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17844979

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017844979

Country of ref document: EP

Effective date: 20190107

NENP Non-entry into the national phase

Ref country code: DE