WO2020156355A1 - Système, dispositif et procédé de partage de charge, carte unique et support de stockage - Google Patents

Système, dispositif et procédé de partage de charge, carte unique et support de stockage Download PDF

Info

Publication number
WO2020156355A1
WO2020156355A1 PCT/CN2020/073377 CN2020073377W WO2020156355A1 WO 2020156355 A1 WO2020156355 A1 WO 2020156355A1 CN 2020073377 W CN2020073377 W CN 2020073377W WO 2020156355 A1 WO2020156355 A1 WO 2020156355A1
Authority
WO
WIPO (PCT)
Prior art keywords
message
single board
board
bum
bum message
Prior art date
Application number
PCT/CN2020/073377
Other languages
English (en)
Chinese (zh)
Inventor
刘新菊
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2020156355A1 publication Critical patent/WO2020156355A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/08—Load balancing or load distribution

Definitions

  • the present disclosure relates to link aggregation load sharing technologies, for example, to a load sharing method, device, system, single board, and storage medium.
  • the multi-chassislinkaggregationgroup supports the aggregation of Ethernet links of two subracks to form a link aggregation group across subracks, which provides links and subrack-level masters for services. ⁇ protection.
  • the cross-subrack channel protection technology only realizes the cross-subrack active and standby protection of services. That is, under normal circumstances, the services are only forwarded on the main link and the main device.
  • the main node device board detects the client side ( Customer Edge, CE)
  • the equipment board of the master node switches the customer service to the board of the standby node through the cross-subrack aggregation group.
  • the service can be protected in this way, the service can only be forwarded from one of the main link or the backup link, and the service load sharing cannot be realized, and the bandwidth of the cross-subrack aggregation group is not fully utilized.
  • the embodiment of the present invention provides a load sharing method, device, system, single board and storage medium.
  • the embodiment of the present application provides a load sharing method, including:
  • At least one of the first single board and the second single board receives service packets, and the first single board and the second single board are each other as a master and backup board;
  • the link where the first board is located and the chain where the second board is located Load sharing and corresponding forwarding.
  • An embodiment of the present application provides a load sharing system, including a first single board and a second single board that are mutually active and standby single boards, and at least one of the first single board and the second single board is configured to: receive Service message; the link where the first single board is located and the link where the second single board is located are set to: in response to determining that the service message is a non-BUM message, load sharing and corresponding forwarding are performed.
  • the embodiment of the present application provides a single board, including a load sharing module and a message processing module.
  • the load sharing module is configured to receive service messages; the message processing module is configured to respond to determining the service messages.
  • load sharing and corresponding forwarding are performed by the link where the single board is located and the link where the other single board that is the active and standby single board with the single board is located.
  • An embodiment of the present application provides a load sharing device, including a processor and a memory for storing a computer program that can run on the processor.
  • a processor When the processor is used to run the computer program, any implementation of the application is executed.
  • the load sharing method described in the example is described in the example.
  • An embodiment of the present application provides a storage medium in which executable instructions are stored, and the executable instructions are executed by a processor to implement the load sharing method described in any embodiment of the present application.
  • Figure 1 is a schematic diagram of the architecture of a load sharing system in related technologies
  • FIG. 2 is a schematic diagram of the architecture of a load sharing system in an embodiment of the application
  • FIG. 3 is a schematic flowchart of a load sharing method in an embodiment of this application.
  • FIG. 4 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the application using a known unicast service as an example
  • FIG. 5 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the present application using an upstream known unicast service as an example;
  • FIG. 6 is a schematic flowchart of a load sharing method in another embodiment of this application.
  • FIG. 7 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the application, where the main board receives the uplink BUM message as an example;
  • FIG. 8 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the application, taking the standby board receiving the uplink BUM message as an example;
  • FIG. 9 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the application, taking the main board receiving downlink BUM packets as an example;
  • FIG. 10 is a schematic diagram of the forwarding process of the load sharing method in an embodiment of the application, taking the standby board receiving the downlink BUM message as an example;
  • FIG. 11 is a schematic diagram of a forwarding process in which the load sharing method in an embodiment of the application adopts the combination of cross-subrack aggregation group and Ethernet multi-environment protection technology;
  • FIG. 12 is a schematic structural diagram of a load sharing system in another embodiment of the application.
  • FIG. 13 is a schematic structural diagram of a single board in an embodiment of the application.
  • FIG. 14 is a schematic structural diagram of a load sharing device in an embodiment of the application.
  • 15 is a schematic flowchart of a load sharing method in an optional embodiment of the application, taking a known unicast packet scenario as an example;
  • FIG. 16 is a schematic flowchart of a load sharing method using the upstream BUM message scenario as an example in an optional embodiment of the application;
  • FIG. 17 is a schematic flowchart of a load sharing method using the following BUM message scenario as an example in an optional embodiment of the application.
  • MC-LAG also known as a link aggregation group between multiple subracks, refers to the aggregation of two corresponding ports on two identical single boards into a protection group to realize the protection of inter-board ports.
  • One access device can simultaneously access two upstream physical devices through the link bundle formed by the cross-subrack aggregation group.
  • Single boards are combined by subracks as units.
  • a subracks can carry multiple different types of boards, and a network element device can include one or more subracks.
  • BUM message refers to Broadcast broadcast message, Unknown unicast unknown unicast message, Multicast multicast message.
  • Aggregation group ports also known as ethernet channels, refer to combining a group of physical ports as a logical channel, that is, a channel-group.
  • the cross-subrack channel protection technology only realizes the cross-subrack active and standby protection of services, that is, under normal circumstances, services are only forwarded on the main link and the main device.
  • the master node equipment board detects that the client-side transmission quality and link status are abnormal, the board switches the client-side business to the standby node's single board through the cross-subrack aggregation group. Service protection, but the service can only be forwarded from the main or backup link, and the bandwidth of the cross-subrack aggregation group cannot be fully utilized.
  • the inventor of the present application discovered in research that the load sharing protection mode can be set for the cross-sub-rack aggregation group, and the load sharing method of the client side business is realized through the cross-sub-rack aggregation group.
  • the main and backup links work at the same time.
  • rules are set to limit the uplink BUM message capabilities of the primary and backup links, which can avoid BUM messages.
  • the messages form a ring between the main and backup boards, so as to ensure the bandwidth utilization and forwarding capacity of the network during the balanced forwarding process of BUM messages through the two links of the main and backup at the same time.
  • FIG. 2 is a schematic structural diagram of a load sharing system provided by an embodiment of the application.
  • the system includes a first single board and a second single board that are mutually active and standby.
  • the first single board and the second single board that are the main and standby boards each other refer to two boards of the same type that together form a cross-subrack aggregation group.
  • the first single board and the second single board The configuration can be the same.
  • the first board can be optionally configured as the main board and the second board can be configured as the backup board; or the second board can be optionally configured according to the aggregation group configuration information.
  • the board is configured as the main board, and the first board is configured as the backup board.
  • the main board A as the first board and the standby board B as the second board as an example.
  • the single-board interconnection port L1 of the first single board and the single-board interconnection port L1 of the second single board are interconnected.
  • the upstream forwarding port C2 of the first board and the upstream forwarding port C2 of the second board are associated with the aggregation group port C1 on the client side to form a cross-subrack aggregation group.
  • the first board and the second board share the load of incoming services through the cross-subrack aggregation group, so that the incoming service packets (ie, downstream packets) from the client side are balanced by the first and second boards It is forwarded to the network side locally, and the service messages (that is, uplink messages) that enter the network side are forwarded to the client side in a balanced manner by the first single board and the second single board.
  • FIG. 3 is a schematic flowchart of a load sharing method provided by an embodiment of this application, which can be applied to the load sharing system shown in FIG. 1, and the method includes the following steps.
  • Step 101 At least one of a first single board and a second single board receives a service message, and the first single board and the second single board are each other as a master and a backup board.
  • the first single board and the second single board are mutually active and standby single boards refer to that the first single board and the second single board are two single boards of the same type that together form a cross-subrack aggregation group, which can be the master single board.
  • the board is the first single board and the backup single board is the second single board; or the main single board is the second single board and the backup single board is the first single board.
  • At least one of the first single board and the second single board receives service packets, which means that the first single board and the second single board receive service packets separately, or the first single board and the second single board jointly receive service packets. Different actual scenes of the text.
  • Step 103 In response to determining that the service message is a non-BUM message, load sharing and corresponding forwarding are performed by the link where the first board is located and the link where the second board is located.
  • non-BUM messages refer to other types of messages except BUM (that is, broadcast messages, unknown unicast messages, and multicast messages), for example, known unicast messages.
  • Load sharing and corresponding forwarding are performed by the link where the first board is located and the link where the second board is located, that is, the service is formed by the link where the first board is located and the link where the second board is located.
  • the main and backup links are forwarded in a balanced manner at the same time, thus making full use of the link bandwidth on the network.
  • the first and second boards receiving downlink packets from the client side as an example.
  • the downlink packets are known unicast packets, they directly pass through the link and the first board where the first board is located.
  • the link where the second board is located performs load sharing, and after reaching the first board and the second board, forwarding is performed according to the destination Media Access Control (MAC) address.
  • MAC Media Access Control
  • the first and second boards directly Forwarding is performed according to the destination MAC address, load sharing is performed through the link where the first board is located and the link where the second board is located, and forwarded to the client side.
  • At least one of the first single board and the second single board receives a service message, and in response to determining that the service message is a non-BUM message, the link and Load sharing and corresponding forwarding are performed on the link where the second single board is located.
  • the load sharing and forwarding of packets through the first board and the second board solves the problem that the related technology can only be forwarded through one of the links of the active and standby boards, and the link bandwidth cannot be fully utilized. Bandwidth utilization is higher.
  • the load sharing method may further include the following steps.
  • Step 105 In response to determining that the service message is a BUM message, restrict the uplink BUM message capability of at least one of the first single board and the second single board according to a setting rule, by restricting the uplink BUM message The first single board and the second single board after the message capability perform corresponding forwarding.
  • BUM message refers to Broadcast broadcast message, unknown unicast unknown unicast message, and multicast multicast message.
  • the service message is a BUM message
  • the BUM message will be broadcast between the first board and the second board.
  • the downstream BUM message Take the downstream BUM message from the client side as an example, the downstream BUM of the first board The message may return to the client side again through the second board, causing the BUM message to form a loop between the first board, the second board and the client side.
  • Restricting the uplink BUM message capability of at least one of the first board and the second board according to the set rules may mean prohibiting the BUM of the first board or the second board according to the forwarding situation of the BUM message Message reporting capability, or limit the reporting capability of the first board and the second board to the BUM message forwarded by the opposite end, by restricting the upstream BUM message capability of the first board and the second board
  • the single board performs corresponding forwarding, so as to avoid the problem of BUM packets forming a loop between the first single board, the second single board, and the client side.
  • the first board and the second board restrict the uplink BUM of at least one of the first board and the second board according to the set rules.
  • the corresponding forwarding can be carried out, so as to avoid the problem of BUM messages forming a loop between the first board, the second board and the client side, and improve the communication between the first board and the second board. Forwarding ability.
  • the step 105 may include the following steps.
  • the second board disables the uplink BUM message function.
  • disabling the upstream BUM message function of the second board refers to prohibiting the upstream forwarding port C2 of the second board from reporting the BUM message.
  • the second board forwards the received upstream BUM packets from the board interconnection port L1 to For the first board, the first board reports the received uplink BUM message to the client side.
  • FIG. 7 Take the first board receiving the upstream packet from the network side as an example.
  • the upstream packet is a BUM packet
  • the first board will receive the upstream BUM packet through the upstream forwarding port C2. Normal forwarding, but when the BUM message passes through the upstream forwarding port C2 of the second board, it will be blocked, thereby ensuring that the upstream BUM message can only be reported to the client side through the upstream forwarding port C2 of the first board.
  • Figure 8 take the second board receiving the upstream packet from the network side as an example.
  • the upstream packet is a BUM packet
  • the second board receives the upstream BUM packet directly and forwards it normally.
  • BUM packets pass through the upstream forwarding port C2 of the second board, they will be blocked.
  • the BUM packets that need to be upstreamed to the client side can only be forwarded through the upstream forwarding port C2 of the first board. In this way, the first board can be guaranteed. Only one node of the single board and the second single board forwards the upstream BUM message to the client side at the same time to prevent the upstream BUM message from forming a loop.
  • the service message is an uplink BUM message
  • restricting the uplink BUM message capability of at least one of the first board and the second board according to the setting rule means prohibiting the second board.
  • the upstream forwarding port C2 of the single board is capable of reporting BUM messages, so as to ensure that only one node on the first and second boards at the same time forwards upstream BUM messages to the client side, preventing upstream BUM messages from forming a loop.
  • the reporting of the received uplink BUM message from the first board to the client side may include the following steps.
  • the second single board adds a designated tag to the received uplink BUM message to obtain an uplink forwarded BUM message, and forwards the uplink forwarded BUM message to the first single board.
  • the first single board reports the received uplink BUM message and the received uplink forwarded BUM message forwarded by the second single board to the client side.
  • the designated label can refer to the virtual local area network (Virtual Local Area Network, vlan) mark set between the first board and the second board in the cross-subrack aggregation group.
  • the vlan mark is added in front of the BUM message to facilitate The first board distinguishes the upstream forwarded BUM message forwarded by the second board from the upstream BUM message received by the local end.
  • the second board disables the upstream BUM message function.
  • the second board adds the vlan tag to the upstream BUM message received to form an upstream forwarding BUM message and forwards it to the first board.
  • the board forwards the upstream BUM message and the upstream BUM message received by the local end to the client side, so as to ensure that the upstream BUM message can only be sent to the client side through the upstream forwarding port C2 of the first board to avoid upstream BUM. Messages are looped.
  • the reporting of the received uplink BUM message from the first board to the client side may include the following steps.
  • the second single board When the link where the first single board is located fails, the second single board enables the uplink BUM message function.
  • the second single board reports the received uplink BUM message and the received uplink forwarded BUM message forwarded by the first single board to the client side.
  • the failure of the link where the first board is located refers to a situation where the link where the first board is located cannot perform normal service forwarding.
  • the second board will switch from disabling the upstream BUM message function to enabling the upstream BUM message function, and switch the business of the link where the first board is located to the second board
  • the second single board reports the received uplink BUM message and the received uplink forwarded BUM message forwarded by the first single board to the client side, thus realizing the service protection function.
  • the load sharing and forwarding mode of the first board and the second board is adopted to realize the balanced sharing and forwarding of service protection.
  • the step 105 may include the following steps.
  • the first single board and the second single board report the uplink BUM message received by the local end to the client side, and forward the received opposite end The upstream forwarded BUM packet is discarded.
  • restricting the uplink BUM message capability of at least one of the first board and the second board according to the setting rule refers to restricting the BUM message forwarded by the first board and the second board to the opposite end Reporting ability.
  • the first board reports the upstream BUM message received by the local end to the client side, and the upstream forwarded BUM message forwarded by the received second board passes through the upstream forwarding port C2 is discarded.
  • the second board reports the upstream BUM message received by the local end to the client side, and discards the upstream forwarded BUM message forwarded by the first board when passing through the upstream forwarding port C2. It can be ensured that the same upstream BUM message can only be sent to the client side through the upstream forwarding port C2 of one of the first board and the second board, so as to prevent the upstream BUM message from forming a loop.
  • the first single board and the second single board report the uplink BUM message received by the local end to the client side, and discard the received uplink forwarded BUM message forwarded by the opposite end. Including the following steps.
  • the first board reports the uplink BUM message received by the local end to the client side, adds the first label to the received uplink BUM message to obtain the uplink forwarded BUM message, and forwards the uplink BUM message Forward to the second board.
  • the second board reports the uplink BUM message received by the local end to the client side, adds the second label to the received uplink BUM message to obtain the uplink forwarded BUM message, and forwards the uplink BUM message Forward to the first board.
  • the first single board and the second single board respectively discard the received uplink forwarding BUM messages forwarded by the opposite end at their respective uplink forwarding ports.
  • the first board adds the first label to the received upstream BUM message and forwards it to The second board.
  • the second board adds a second label to the received upstream BUM message and forwards it to The first board.
  • the first board discards the received upstream forwarding BUM packet forwarded by the second board when it passes through the upstream forwarding port C2, and the second board discards the upstream forwarding BUM packet received by the first board forwarded by the upstream Discard when forwarding port C2. In this way, it can be ensured that the same upstream BUM message can only be sent to the client side through the upstream forwarding port C2 of one of the first board and the second board, so as to prevent the upstream BUM message from forming a loop.
  • the step 105 may include the following steps.
  • the second board disables the uplink BUM message function.
  • the second board In response to determining that the service message is a downlink BUM message, the second board adds a designated tag to the received downlink BUM message to obtain a downlink forwarded BUM message, and forwards the downlink forwarded BUM message to all For the first single board, the first single board discards the received downstream forwarding BUM message forwarded by the second single board at the upstream forwarding port of the first single board.
  • restricting the uplink BUM message capability of at least one of the first single board and the second single board according to the setting rule refers to prohibiting the uplink forwarding port C2 of the second single board from reporting the BUM message capability.
  • the designated label can refer to the VLAN tag set between the first board and the second board in the cross-subrack aggregation group.
  • FIG. 9 Take the first board receiving the downstream message from the customer side (CE) as an example.
  • the downstream message is a BUM message
  • the first board directly performs normal forwarding, and the first board will The received downstream BUM message is forwarded to the second board.
  • the downstream BUM message passes through the upstream forwarding port C2 of the second board, it will be blocked. This ensures that the downstream BUM message received from the first board The message will not be uploaded and forwarded back to the CE, avoiding the looping of downstream BUM messages.
  • Figure 10 taking the second board receiving the downstream message from the client side as an example.
  • the second board When the downstream message is a BUM message, the second board performs normal forwarding and adds a designated label to the downstream BUM message Then, it is forwarded to the first board through the board interconnection port L1. After the first board receives the downstream forwarded BUM message with the specified label, it forwards the downstream forwarded BUM message to the upstream forwarding port C2. The downstream BUM packets are forwarded normally through the upstream forwarding port C2 of the first board. This ensures that the downstream BUM packets received from the second board will not be forwarded and forwarded back to the CE , To avoid the loop of downlink BUM messages.
  • the second board when at least one of the first board and the second board receives a service message and determines that the service message is a downlink BUM message entered by the client, the second board will receive the downlink BUM The message is forwarded to the network side, and the received downstream BUM message is added with a specified label to form a downstream forwarded BUM message, and then forwarded to the first board through the board interconnection port L1, and the downstream forwarded BUM message passes through the first board When the upstream forwarding port C2 is discarded. In this way, the downlink BUM message can be prevented from returning to the client side again via the uplink forwarding port C2 of the first board, and the loop of the downlink BUM message can be avoided.
  • the step 105 may include the following steps.
  • the first single board and the second single board respectively forward the received downlink BUM message to the network side, and forward the received downlink BUM message to the network side.
  • Downlink forwarding BUM packets are discarded at their respective upstream forwarding ports.
  • restricting the uplink BUM message capability of at least one of the first board and the second board according to the setting rule refers to restricting the BUM message forwarded by the first board and the second board to the opposite end Reporting ability.
  • the first board forwards the downlink BUM packet received at the local end to the network side, and the received downlink forwarded BUM packet forwarded by the second board passes the uplink forwarding port
  • the second single board forwards the downstream BUM message received by the local end to the network side, and discards the received downstream forwarded BUM message forwarded by the first single board when passing through the upstream forwarding port C2.
  • the step of discarding the received downlink forwarding BUM message of the opposite end at the respective uplink forwarding port may include the following steps.
  • the first single board adds a first label to the received downlink BUM message to obtain a downlink forwarded BUM message, and forwards the downlink forwarded BUM message to the second single board, and the second single board
  • the received downstream forwarding BUM message forwarded by the first single board is discarded at the upstream forwarding port of the second single board.
  • the second single board adds a second label to the received downlink BUM message to obtain a downlink forwarded BUM message, and forwards the downlink forwarded BUM message to the first single board, and the first single board will The received downstream forwarding BUM message forwarded by the second single board is discarded at the upstream forwarding port of the first single board.
  • the first board adds the first label to the received downstream BUM message and forwards it to The second board.
  • the second board adds a second label to the received downstream BUM packets and forwards them to The first board.
  • the first board discards the received downstream forwarded BUM packets forwarded by the second board when passing through the upstream forwarding port C2
  • the second board discards the received downstream forwarded BUM packets forwarded by the first board through the upstream Discard when forwarding port C2.
  • step 101 before at least one of the first board and the second board receives a service packet, the method may further include the following steps.
  • Step 1021 Obtain aggregation group configuration information.
  • Step 1022 According to the aggregation group configuration information, respectively associate the upstream forwarding port of the first board with the client-side aggregation group port, and the upstream forwarding port of the second board with the client-side aggregation group port.
  • the ports are associated to form a cross-subrack aggregation group, and the cross-subrack aggregation group is configured to perform load sharing through the associated aggregation group ports.
  • the aggregation group configuration information can be configured on the management side of the network.
  • two cascaded boards or devices can be configured as active and standby, and the uplink forwarding ports of the two boards can be associated with the aggregation group ports on the client side to form a cross-subrack aggregation group.
  • Obtaining aggregation group configuration information can also include: creating a vlan mark between the active and standby nodes, marking forwarding BUM packets, and blocking the associated aggregation group port of the standby board, that is, the upstream BUM forwarding function of the upstream forwarding port C2, to ensure Only one of the active and standby nodes can forward the BUM message upstream.
  • the active and standby links where the two single boards are located can perform load sharing and forwarding according to the set rules, thereby improving link bandwidth utilization and service forwarding capabilities.
  • the load sharing method may further include the following steps.
  • the master node is determined according to the master node selection instruction, and the link between the master node and the nodes adjacent to the master node is set as a low-power lossy network routing protocol (Routing Protocol for Low Power and Lossy Network) (LLN), RPL) link.
  • LLC Low Power and Lossy Network
  • ring network protection is established on the first single board, the second single board, and the third single board and the fourth single board communicatively connected to the first single board and the second single board.
  • the master node is selected and determined on the ring network, and the link between the master node and the node adjacent to the master node is set to RPL link.
  • the link is not faulty, in order to ensure that the business is not ringed, the RPL link is in a blocked state, and the link is detected by the Connectivity Fault Management (cfm).
  • Connectivity Fault Management cfm
  • the RPL link is set to the forwarding state, the ring network is switched, and the ring network protection is generated.
  • a ring network protection is established between the third board C and the fourth board D and the main board A and the backup board B.
  • select the fourth board D as the master node, and set the link between the fourth board D and the standby board B as the RPL link.
  • the RPL link is in a blocked state to ensure that the business does not form a loop.
  • Use cfm to detect the link.
  • the RPL link is set to the forwarding state, the ring network is switched, and the ring network protection is generated.
  • MC-LAG and Ethernet Ring Protection Switching are combined to implement a load sharing method to ensure that services are not looped in the ring network architecture.
  • services Load sharing is carried out across subrack aggregation groups, and forwarding is performed on the main and backup links at the same time, which greatly improves bandwidth utilization and forwarding capabilities.
  • the cross-subrack aggregation group will promptly switch the service of the failed link to another normal link, thereby achieving service protection.
  • the failed link recovers, the service continues to perform load sharing through the active and standby links.
  • An embodiment of the present application provides a load sharing system, including a first single board and a second single board that are mutually active and standby single boards, and at least one of the first single board and the second single board is configured to: receive Service message; the link where the first single board is located and the link where the second single board is located are set to: in response to determining that the service message is a non-BUM message, load sharing and corresponding forwarding are performed.
  • the first board and the second board which are the main and backup boards for each other, mean that the first board is the main board and the second board is the backup board; or the first board is the backup board and the second board is the backup board.
  • the board is the main board. Refer to Figure 12, taking the first board as the main board A and the second board as the backup board B as an example, the link where the first board is located is the main link, and the link where the second board is located is the backup link
  • the upstream forwarding port C2 of the first single board and the upstream forwarding port C2 of the second single board are associated with the aggregation group port C1 on the client side to form a cross-subrack aggregation group.
  • the first single board and the second single board identify the type of service message, and when it is determined that the service message is a non-BUM message, the link and The link where the second single board is located performs load sharing and corresponding forwarding, so that the first single board and the second single board complete cross-board aggregation service forwarding, and the primary link and the backup link jointly implement the load sharing of the active and standby links.
  • At least one of the first single board and the second single board is further configured to: in response to determining that the service message is a BUM message, restrict the respective uplink BUM message according to a setting rule Ability, to limit the uplink BUM message capability and forward accordingly.
  • the second board adds a vlan tag to the received upstream BUM message and forwards it to the first board so that the first board can recognize the upstream forwarded BUM message forwarded by the second board.
  • the second board is set to disable the uplink BUM message function.
  • the first single board and the second single board are set to: in response to determining that the service packet is an uplink BUM packet, report the received uplink BUM packet from the first single board to the client side .
  • the second board is configured to: disable the uplink BUM message function; in response to determining that the service message is a downlink BUM message, add a designated tag to the received downlink BUM message Obtain a downlink forwarded BUM message, and forward the downlink forwarded BUM message to the first board.
  • the first single board is configured to discard the received downlink forwarding BUM message forwarded by the second single board at the uplink forwarding port of the first single board.
  • the second board is further configured to enable the uplink BUM message function when the link where the first board is located fails.
  • the first single board and the second single board are configured as:
  • the downlink BUM message received by the local end is forwarded to the network side, and the downlink forwarded BUM message forwarded by the opposite end is received at the respective uplink forwarding port throw away.
  • restricting the upstream BUM message capability of at least one of the first board and the second board according to the setting rule refers to restricting the upstream forwarding port pair of the first board and the second board to receive Reporting capability of BUM packets forwarded by the peer.
  • the first board adds the first label to the BUM message and forwards it to the second board, and the second board sends the BUM message to the second board.
  • the message is forwarded to the first board after adding the second label.
  • the first label and the second label can be different vlan labels, so that the first board and the second board can identify the received BUM message forwarded by the opposite end Then discard it at the upstream forwarding port C2.
  • the first single board and the second single board are further configured to: obtain aggregation group configuration information; according to the aggregation group configuration information, the first single board transfers the first single board
  • the uplink forwarding port of the second single board is associated with the aggregation group port on the client side, and the second single board associates the uplink forwarding port of the second single board with the aggregation group port to form a cross-subrack aggregation group.
  • the rack aggregation group is configured to perform load sharing through the associated aggregation group ports.
  • the load sharing system determines that the service packet is a non-BUM packet through the first board and the second board
  • the link between the first board and the chain where the second board is located Load sharing and corresponding forwarding.
  • the first single board and the second single board forward packets through load sharing, which solves the problem that the related technology can only be forwarded through one of the links of the active and standby single boards, thereby achieving higher bandwidth utilization.
  • the first board and the second board determine that the service message is a BUM message
  • the first board and the second board restrict at least one of the first board and the second board according to the set rule
  • the upstream BUM message capability is then forwarded accordingly. In this way, the problem of BUM messages forming a loop between the first board, the second board and the client side can be avoided, and the first board and the second board can be improved. 2.
  • the forwarding capability of the single board is a non-BUM packet through the first board and the second board.
  • an embodiment of the present invention also provides a single board, including a load sharing module 21 and a message processing module 23.
  • the load sharing module 21 is configured to receive service messages; the message processing module 23, In response to determining that the service message is a non-BUM message, load sharing and corresponding forwarding are performed by the link where the single board is located and the link where the other single board that is the master and backup board is mutually active with the single board .
  • the single board may refer to the first single board described in any embodiment of the present application, and the other single board that is the active and standby single board with the single board refers to any implementation of the present application
  • the other board refers to the first board described in any embodiment of the present application, and is not limited herein.
  • the message processing module 23 is further configured to, in response to determining that the service message is a BUM message, limit the uplink BUM message capability of the board according to a setting rule, and based on the restriction The single board after the uplink BUM message capability and the other single board perform corresponding forwarding.
  • the message processing module 23 is further configured to prohibit the upstream BUM message function of the upstream forwarding port of the board; in response to determining that the service message is an upstream BUM message, it will receive Add a specified label to the upstream BUM message to obtain the upstream forwarded BUM message, and forward the upstream forwarded BUM message to the other single board, and the other single board forwards the upstream forwarded BUM message to the Report to the client at the upstream forwarding port of the other board; or, in response to determining that the service message is a downstream BUM message, add the designated label to the received downstream BUM message to obtain a downstream forwarded BUM message And forward the downstream forwarded BUM message to the another single board, and the other single board discards the downstream forwarded BUM message at the upstream forwarding port of the other single board.
  • the message processing module 23 is further configured to enable the uplink BUM message function of the uplink forwarding port of the single board when the link where the other single board is located fails.
  • the message processing module 23 is further configured to, in response to determining that the service message is an uplink BUM message, report the received uplink BUM message to the client side, and send the received uplink BUM message. Adding a specified label to the message to obtain an upstream forwarding BUM message, forwarding the upstream forwarding BUM message to the other single board, and discarding the received upstream forwarding BUM message forwarded by the other single board; Or, in response to determining that the service message is a downlink BUM message, forward the received downlink BUM message to the network side, add a designated label to the received downlink BUM message to obtain a downlink forwarded BUM message, and transfer all The downstream forwarded BUM message is forwarded to the other single board, and the received downstream forwarded BUM message of the other single board is discarded at the upstream forwarding port of the single board.
  • the load sharing module 21 is further configured to obtain aggregation group configuration information, associate the uplink forwarding port of the board with the aggregation group port on the client side according to the aggregation group configuration information, and form a cross-connect Subrack aggregation group.
  • the single board provided in the above embodiment implements the load sharing method
  • only the division of the above multiple program modules is used as an example.
  • the above steps can be allocated to different program modules to complete according to needs, that is, The internal structure of the program for the single board to implement the load sharing method is divided into different program modules to complete all or part of the processing described above.
  • the message processing module 23 may include a BUM message processing module and a message forwarding module.
  • the BUM message processing module includes a tag creation unit, a message judgment unit, and a BUM message processing unit.
  • the tag creation unit It is set to create a special tag vlan between the active and standby boards, which is used to identify the BUM packets transmitted between the active and standby boards.
  • the message judging unit is set to judge the received messages by the active and standby nodes, and distinguish between known unicast and BUM messages.
  • the BUM message processing unit is configured to turn off the upstream BUM function of the upstream forwarding port of the backup board, the downstream BUM message received by the main board is forwarded normally, and the downstream BUM message received by the backup board is marked
  • the vlan label is forwarded to the main single board for forwarding.
  • the main single board directly forwards it normally after receiving it, and the standby single board also forwards it normally after receiving it, but the upstream forwarding port of the standby single board does not forward it.
  • the message forwarding module includes a downlink known unicast message forwarding unit, a downlink BUM message forwarding unit, an uplink known unicast message forwarding unit, and an uplink BUM message forwarding unit.
  • the known downlink unicast message forwarding unit is set to directly perform the forwarding according to the destination MAC address after load sharing through the active and standby links.
  • the downlink BUM message forwarding unit is configured to directly forward the downlink BUM message received by the main single board, and the uplink forwarding port of the standby single board does not forward the BUM message; Downlink BUM packets are forwarded to the main single board after adding a vlan label, and other ports perform normal forwarding. After the main single board receives a vlan-labeled packet, it does not forward to the upstream forwarding port, but other ports perform normal forwarding. .
  • the uplink known unicast message forwarding unit is configured to forward the known uplink unicast messages directly according to the destination MAC address after load sharing through the active and standby links.
  • the upstream BUM message forwarding unit is set to perform normal forwarding directly after receiving the upstream BUM message on the active and standby single board, and the upstream forwarding port of the standby single board does not perform upload forwarding.
  • the embodiment of the present invention also provides a load sharing device.
  • the load sharing device may belong to one or more single boards in the network. Please refer to FIG. 14.
  • the load sharing device includes a processor 201 and a processor for storing 201 is a storage medium 202 of a computer program running on it.
  • the processor 201 is configured to execute the steps of the load sharing method provided in any embodiment of the present application when running the computer program.
  • the processor 201 and the storage medium 202 do not mean that the corresponding number is one, but may be one or more.
  • the storage medium 202 may store an operating system and a load sharing device for implementing the load sharing method provided in the embodiment of the present invention.
  • the processor 201 is used to improve calculation and control capabilities and support the operation of the entire server.
  • the embodiment of the present invention also provides a storage medium, for example, including a memory storing a computer program, and the computer program can be executed by a processor to complete the steps of the load sharing method provided in any embodiment of the present application.
  • the storage medium can be Ferroelectric Random Access Memory (FRAM), Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), erasable and programmable Read-only memory (Erasable Programmable Read-Only Memory, EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory (Flash Memory), magnetic surface memory, optical disk, or optical disk only CD-ROM (Compact Disc Read-Only Memory) and other memories; it can also be a variety of devices including one or any combination of the foregoing memories.
  • FRAM Ferroelectric Random Access Memory
  • ROM Read-Only Memory
  • PROM Programmable Read-Only Memory
  • EPROM Erasable Programmable Read-Only Memory
  • EEPROM Electrically Erasable Programmable Read
  • At least one of the first single board and the second single board receives a service packet, and responds to determining that the service packet is a non-BUM packet
  • load sharing and corresponding forwarding are performed by the link where the first single board is located and the link where the second single board is located. In this way, the load sharing and forwarding of packets by the first single board and the second single board solves the problem that the related technology can only be forwarded through one of the links of the active and standby single boards, thereby achieving higher bandwidth utilization.
  • the load sharing methods in different scenarios of known unicast packet, uplink BUM packet, and downlink BUM packet transmission are exemplified below.
  • the load sharing method includes the following steps.
  • Step S11 at least one of the main single board and the standby single board receives a known unicast message.
  • Step S12 Determine that the known unicast message is an uplink message or a downlink message.
  • steps S131 to S133 are executed; when it is determined to be an uplink known unicast message, steps S141 to S143 are executed.
  • Step S131 When it is determined that it is a known downlink unicast message, the active and standby links perform load sharing and forward them to the active and standby boards.
  • step S132 the active and standby single boards forward according to the destination MAC address of the known unicast message.
  • Step S133 When one of the active and standby links fails, the service of the failed link is switched to the normal link for forwarding.
  • Step S141 When it is determined that it is an uplink known unicast message, the active and standby single boards forward it according to the destination MAC address.
  • Step S142 the services that reach the client side are load-shared by the active and standby links.
  • Step S143 When one of the active and standby links fails, the service of the failed link is switched to the normal link for forwarding.
  • the active and standby single boards determine that the service message is a known unicast message
  • the active and standby links perform load sharing and corresponding forwarding.
  • the active and standby boards are used to forward packets through load sharing, which solves the problem that the related technology can only forward through one of the links of the active and standby boards but cannot effectively use the network bandwidth, and the bandwidth utilization is higher.
  • the load sharing method further includes the steps of acquiring configuration information, and forming a cross-subrack aggregation group according to the configuration information.
  • the acquiring configuration information and forming a cross-subrack aggregation group based on the configuration information may include: creating an aggregation group port on the client-side device CE to achieve service load sharing; configuring the aggregation group port on the CE device, which can be configured as static or as required Manual aggregation group; establish MC-LAG on the first board and the second board, board A is the main board, board B is the backup board, and the upstream forwarding port C2 of the main board A and the backup board B
  • port L1 is the board interconnection port, and the protection mode selects load sharing; the upstream forwarding port C2 on the standby board prohibits the BUM message sending function; the main board A and Configure the designated vlan label between the standby single
  • the load sharing method may include the following steps.
  • Step S21 at least one of the main single board and the standby single board receives an uplink BUM message.
  • Step S22 Determine that the uplink BUM message is received by the master board or the backup board.
  • step S231 When the master board receives the upstream BUM message, step S231 is executed; when the backup board receives the upstream BUM message, steps S241 to S242 are executed.
  • step S231 the main single board directly forwards the received uplink BUM message normally.
  • step S241 the standby single board prohibits the ability of its upstream forwarding port to report the BUM message, and normally forwards the received upstream BUM message.
  • step S242 the standby single board tags and forwards the received uplink BUM message to the main single board, and reports to the client side through the uplink forwarding port of the main single board.
  • the standby board prevents the upstream forwarding port from reporting BUM messages to ensure that there is only one node in the active and standby boards at the same time Forwarding the upstream BUM message to the client side prevents the upstream BUM message from forming a loop.
  • the forwarding capability of the active and standby boards is improved. It should be noted that in this embodiment, restricting the upstream BUM message capability of at least one of the first board and the second board according to the setting rules is to prohibit the upstream forwarding port C2 of the standby board. Take the reporting capability of BUM packets as an example.
  • the backup board By disabling the reporting capability of BUM packets on the upstream forwarding port C2 of the backup board, the backup board does not need to perform the load sharing and forwarding of upstream BUM packets.
  • the upstream BUM messages received by the single board are distinguished, and the main single board can forward it normally. In this way, the first single board and the second single board can greatly simplify the way of forwarding BUM packets and avoid the problem of BUM packets forming a loop.
  • the load sharing method may include the following steps.
  • Step S31 at least one of the main single board and the standby single board receives the downlink BUM message.
  • Step S22 It is determined that the downlink BUM message is received by the master board or the backup board.
  • steps S331 to S332 are executed; when the slave single board receives the downlink BUM message, steps S341 to S342 are executed.
  • Step S331 When the main single board receives the downlink BUM message, the main single board directly forwards the received downlink BUM message normally.
  • step S332 the main single board normally forwards the downlink BUM message carrying the vlan tag and forwarded by the main single board on other ports except the uplink forwarding port C2.
  • Step S341 The standby single board forwards the received downlink BUM message on other ports except the single board interconnection port L1 normally, and adds the vlan tag to the downstream BUM message and forwards it to the single board interconnection port L1. Main board.
  • step S342 the standby single board prohibits its upstream forwarding port C2 from reporting the BUM message, and the received downstream BUM message forwarded by the primary single board is discarded at its upstream forwarding port C2.
  • the standby board prevents the upstream forwarding port from reporting the BUM message to ensure that the standby board is forwarded to the main board.
  • Downlink BUM packets will not be forwarded and sent back to the client side through the upstream forwarding port of the main board to prevent the downstream BUM packets from forming a loop.
  • the forwarding of the main and standby boards is improved. ability. It should be noted that in this embodiment, restricting the upstream BUM message capability of at least one of the first board and the second board according to the setting rules is to prohibit the upstream forwarding port C2 of the standby board.
  • the backup board can only add designated labels for the downstream BUM packets and forward them To the main single board, and the main single board recognizes the downstream BUM packet forwarded by the standby single board and discards it at the upstream forwarding port C2. In other scenarios, the main single board can forward it normally. In this way, the first single board and the second single board can greatly simplify the way of forwarding BUM packets and avoid the problem of BUM packets forming a loop.

Abstract

La présente invention concerne un système, un dispositif et un procédé de partage de charge, une carte unique et un support de stockage, le procédé comprenant les étapes suivantes : une première carte unique et/ou une seconde carte unique reçoit un message de service et la première carte unique et la seconde carte unique sont des cartes uniques mutuellement en veille et actives ; en réponse à la détermination du fait que le message de service est un message non BUM, un partage de charge et un transfert correspondant sont effectués par la liaison où est située la première carte unique et par la liaison où est située la seconde carte unique.
PCT/CN2020/073377 2019-01-29 2020-01-21 Système, dispositif et procédé de partage de charge, carte unique et support de stockage WO2020156355A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910085733.2 2019-01-29
CN201910085733.2A CN111491334B (zh) 2019-01-29 2019-01-29 负载分担方法、装置、系统、单板及存储介质

Publications (1)

Publication Number Publication Date
WO2020156355A1 true WO2020156355A1 (fr) 2020-08-06

Family

ID=71812162

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/073377 WO2020156355A1 (fr) 2019-01-29 2020-01-21 Système, dispositif et procédé de partage de charge, carte unique et support de stockage

Country Status (2)

Country Link
CN (1) CN111491334B (fr)
WO (1) WO2020156355A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112291160B (zh) * 2020-09-23 2022-03-01 浪潮思科网络科技有限公司 一种bum报文抑制方法、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101242335A (zh) * 2008-02-26 2008-08-13 中兴通讯股份有限公司 多链路负荷分担的处理装置、处理方法及mpls vpn网络设备
CN101252531A (zh) * 2008-04-02 2008-08-27 杭州华三通信技术有限公司 一种实现负载分担和主备倒换的方法、系统及设备
CN101394260A (zh) * 2007-09-17 2009-03-25 华为技术有限公司 一种实现主备倒换及负荷分担的方法和装置
WO2011113395A2 (fr) * 2011-04-27 2011-09-22 华为技术有限公司 Procédé et appareil d'équilibrage des charges
CN108134744A (zh) * 2017-12-06 2018-06-08 新华三技术有限公司 一种irf堆叠中的负载分担方法和单板

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100440799C (zh) * 2005-09-06 2008-12-03 华为技术有限公司 一种接入设备中主备板备份及倒换的方法
CN101132374B (zh) * 2006-10-09 2012-02-08 杭州华三通信技术有限公司 一种发送报文的方法及装置
CN101674240B (zh) * 2009-09-24 2012-05-09 中兴通讯股份有限公司 用于分布式系统的单节点双归的实现方法和装置
WO2013075284A1 (fr) * 2011-11-22 2013-05-30 华为技术有限公司 Procédé et dispositif pour la mise en œuvre d'un groupe de ressources en bande de base lte
CN104426794B (zh) * 2013-08-23 2018-06-26 华为技术有限公司 一种报文转发方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101394260A (zh) * 2007-09-17 2009-03-25 华为技术有限公司 一种实现主备倒换及负荷分担的方法和装置
CN101242335A (zh) * 2008-02-26 2008-08-13 中兴通讯股份有限公司 多链路负荷分担的处理装置、处理方法及mpls vpn网络设备
CN101252531A (zh) * 2008-04-02 2008-08-27 杭州华三通信技术有限公司 一种实现负载分担和主备倒换的方法、系统及设备
WO2011113395A2 (fr) * 2011-04-27 2011-09-22 华为技术有限公司 Procédé et appareil d'équilibrage des charges
CN108134744A (zh) * 2017-12-06 2018-06-08 新华三技术有限公司 一种irf堆叠中的负载分担方法和单板

Also Published As

Publication number Publication date
CN111491334B (zh) 2021-05-25
CN111491334A (zh) 2020-08-04

Similar Documents

Publication Publication Date Title
US11178001B2 (en) Multi-stage switch fabric fault detection and handling
US9660939B2 (en) Protection switching over a virtual link aggregation
US7061875B1 (en) Spanning tree loop guard
US7751329B2 (en) Providing an abstraction layer in a cluster switch that includes plural switches
US7639605B2 (en) System and method for detecting and recovering from virtual switch link failures
US8520508B2 (en) Spanning tree ring protocol
US8411690B2 (en) Preventing data traffic connectivity between endpoints of a network segment
CN101155109B (zh) 一种以太网交换系统及设备
US8649379B2 (en) Method and apparatus for configuring a link aggregation group on a stacked switch
US9485198B1 (en) Methods and apparatus for multicast traffic failover in a network
US11509494B2 (en) Protocol independent multicast (PIM) designated router (DR) election
US9984028B2 (en) Redundancy for port extender chains
US11979283B2 (en) Stacking-port configuration using zero-touch provisioning
WO2017113929A1 (fr) Procédé et contrôleur de commande de grappe de dispositifs de transfert dans un réseautage défini par logiciel
CN111901133A (zh) 一种组播切换方法、装置、网络设备和存储介质
US20050163137A1 (en) Switching mesh with broadcast path redundancy
CN114401191B (zh) 错误配置的上行链路标识
WO2020156355A1 (fr) Système, dispositif et procédé de partage de charge, carte unique et support de stockage
US11296980B2 (en) Multicast transmissions management
US10567218B2 (en) Systems and methods for enabling frontside stacking of switches
US11985002B2 (en) Protocol independent multicast (PIM) designated router (DR) election

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: 20747674

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20747674

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 04/02/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20747674

Country of ref document: EP

Kind code of ref document: A1