WO2023103461A1 - 基于clos架构的报文跨板组播复制转发方法和系统 - Google Patents

基于clos架构的报文跨板组播复制转发方法和系统 Download PDF

Info

Publication number
WO2023103461A1
WO2023103461A1 PCT/CN2022/114805 CN2022114805W WO2023103461A1 WO 2023103461 A1 WO2023103461 A1 WO 2023103461A1 CN 2022114805 W CN2022114805 W CN 2022114805W WO 2023103461 A1 WO2023103461 A1 WO 2023103461A1
Authority
WO
WIPO (PCT)
Prior art keywords
line card
message
service line
private
port
Prior art date
Application number
PCT/CN2022/114805
Other languages
English (en)
French (fr)
Inventor
郑成龙
谢永恒
万月亮
Original Assignee
北京锐安科技有限公司
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 北京锐安科技有限公司 filed Critical 北京锐安科技有限公司
Publication of WO2023103461A1 publication Critical patent/WO2023103461A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4633—Interconnection of networks using encapsulation techniques, e.g. tunneling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/20—Support for services
    • H04L49/201—Multicast operation; Broadcast operation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/25—Routing or path finding in a switch fabric
    • H04L49/252—Store and forward routing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

  • the present disclosure relates to the technical field of communication, for example, to a method and system for cross-board multicast replication and forwarding of messages based on CLOS architecture.
  • the CLOS architecture defines a geometric topology structure, which mainly consists of the main control card, line-card Card (Fabric-card) composition.
  • the CLOS architecture completely separates the control plane and the data plane.
  • the switching line card serves as the forwarding center of the system, which greatly improves the system forwarding capability; on the other hand, it greatly improves the reliability of the forwarding plane and avoids Impact of faults on the forwarding plane.
  • the original traffic is distributed to a certain logic chip according to the equalization algorithm of the same source and same destination. After the logic chip completes the relevant business, it needs to share information with other business line cards, usually by means of multicast of the switching chip in the system.
  • the forwarding function is implemented in the form of private packets.
  • the backplane bandwidth resources are responsible for the traffic forwarding of the entire system.
  • the multicast forwarding of private packets requires additional backplane bandwidth resources.
  • the embodiment of the present application provides a method and system for cross-board multicast copying and forwarding of messages based on the CLOS architecture, so as to realize the function of private message multicast forwarding under the CLOS architecture.
  • the embodiment of the present application provides a method for copying and forwarding message cross-board multicast based on CLOS architecture, the method comprising:
  • the switching chip of the first service line card filters the original message to obtain the message to be processed, and sends the message to be processed to the first advanced logic processing chip of the first service line card according to the same-source and same-destination balancing method conduct business processing;
  • the first advanced logic processing chip After the first advanced logic processing chip performs business processing on the message to be processed, it determines the private message that needs to be shared, encapsulates the private message, and forwards it to the switching chip of the first service line card ;
  • the switching chip of the first service line card When the switching chip of the first service line card recognizes that the message type of the private message is a private forwarding message, determine the preset selection algorithm that matches the message type of the private forwarding message, according to the A preset selection algorithm selects a port from the ports connected to the first service line card and the switching line card as a target port, and forwards the private message to the switching line card through the target port;
  • the switching line card receives the private message, completes multicast replication, and forwards the copied private message to a target service line card.
  • the embodiment of the present application provides a message cross-board multicast replication and forwarding system based on CLOS architecture, including at least two service line cards and at least one switching line card, and the at least two service line cards each include A switch chip and at least one advanced logic processing chip, the advanced logic processing chips of the at least two service line cards are of different types, each switch line card is connected to all service line cards through a backplane port, and the at least two The business line card includes the first business line card and the target business line card; wherein:
  • the switching chip of the first service line card is configured to filter the original message to obtain the message to be processed, and send the message to be processed to the first service line card in the same-source and same-destination balanced manner.
  • the first advanced logic processing chip for business processing is configured to filter the original message to obtain the message to be processed, and send the message to be processed to the first service line card in the same-source and same-destination balanced manner.
  • the first advanced logic processing chip is configured to determine the private message that needs to be shared after performing business processing on the message to be processed, encapsulate the private message, and forward it to the first service line card switch chip;
  • the switching chip of the first service line card is further configured to determine the preset selection matched by the message type of the private forwarding message when it recognizes that the message type of the private message is a private forwarding message Algorithm, according to the preset selection algorithm, select a port from the ports connected to the first service line card and the switching line card as a target port, and forward the private message to the switching line card through the target port;
  • the switching line card is configured to receive the private message, complete multicast replication, and forward the copied private message to a target service line card.
  • Fig. 1 a is the flow chart of a kind of message cross-board multicast copying and forwarding method based on CLOS framework that the embodiment one of the application provides;
  • FIG. 1b is an overall block diagram of a device system under the CLOS architecture provided by Embodiment 1 of the present application;
  • Fig. 2 is a flow chart of a method for copying and forwarding messages across boards based on CLOS architecture provided in Embodiment 2 of the present application;
  • Fig. 3 is a flow chart of a method for copying and forwarding messages across board multicasts based on the CLOS architecture provided in Embodiment 3 of the present application;
  • FIG. 4 is a schematic diagram of a message cross-board multicast replication and forwarding system based on the CLOS architecture provided in Embodiment 4 of the present application.
  • Figure 1a is a flow chart of a method for multicast copying and forwarding of messages across boards based on CLOS architecture provided by Embodiment 1 of the present application. This embodiment is applicable to the case of multicast copying and forwarding of messages across boards under the CLOS architecture.
  • the method It can be executed by a message cross-board multicast replication and forwarding system based on the CLOS architecture, including the following steps:
  • Step 110 the switching chip of the first service line card filters the original message to obtain the message to be processed, and sends the message to be processed to the first A high-level logic processing chip performs business processing.
  • Fig. 1b is an overall block diagram of the device system under the CLOS architecture. All the control functions of the system are in charge of the main control card, and there is no connection between the service line cards. Each switch line card is connected to all service line cards through backplane ports. Service line cards focus on business processing, and switch line cards focus on traffic forwarding in the system.
  • the service line cards include multiple different types of service line cards or multiple service line cards of the same type, and the service line cards can be implemented by means of switching chips and advanced logic processing chips.
  • the interface between the switching chip and the logic chip is an aggregation port composed of multiple physical links, which is used for redundant sharing of traffic. All switching line cards form a backplane forwarding matrix.
  • Each service line card and switching line card The backplane port between them is a backplane aggregation port composed of multiple physical ports, which is used for redundant sharing of traffic.
  • the switch chip is responsible for traditional basic functions such as aggregation and distribution.
  • the advanced logic processing chip realizes advanced functions such as in-depth analysis of complex messages.
  • the advanced logic processing chip refers to the analysis, processing and forwarding functions of traffic that can be realized through software devices, for example, network processor (Network Processor, NP), field programmable logic gate array (Field Programmable Gate Array, FPGA), MIPS multi-core processor or central processing unit (central processing unit, CPU), etc., can also be Other devices capable of traffic analysis, processing, and forwarding through software.
  • network processor Network Processor, NP
  • field programmable logic gate array Field Programmable Gate Array, FPGA
  • MIPS multi-core processor or central processing unit central processing unit, CPU
  • CPU central processing unit
  • the switching chip filters the original message based on a preset filtering rule in the switching chip to obtain a message to be processed; wherein, the filtering rule may be an Access Control List (Access Control List) Lists, ACL) rules.
  • the filtering rule may be an Access Control List (Access Control List) Lists, ACL) rules.
  • Each switching line card is connected to all service line cards through a backplane port, so the switching chip sends the message to be processed to the The first advanced logic processing chip.
  • the service panel has a panel input port, and before the switch chip filters the original message, the switch chip obtains the input original message through the panel port of the first service line card.
  • the original message enters the system through the panel port of the first service line card, the message enters the switching chip in the service line card, and the original message passes the ACL rule issued by the switching chip in advance , to filter out the pending packets that need to be processed by the advanced logic processing chip TypeA.
  • the switching chips of all service line cards can filter the original packets to obtain the pending packets that need to be processed by the advanced logic processing chip TypeA .
  • the message to be processed is distributed to the advanced logic processing chip TypeA according to the balance algorithm of the same source and the same destination.
  • the message to be processed is sent to For the business line card where the advanced logic processing chip TypeA is located.
  • the original message enters the switching chip of service line card 1 for filtering to obtain the message to be processed, and the message to be processed enters the advanced logic processing chip TypeA of service line card 1 for processing in a balanced manner of the same source and same destination.
  • Step 120 After the first advanced logic processing chip performs business processing on the message to be processed, determine the private message that needs to be shared, encapsulate the private message, and forward it to the first service line card switch chip.
  • the first advanced logic processing chip encapsulates the private message in a set encapsulation format
  • the set encapsulation format includes a private message header and private message information
  • the private message header contains The Media Access Control layer (Media Access Control, MAC) field or the Virtual Local Area Network (Virtual Local Area Network, VLAN) field is set with a private message tag, or the private message header is set with a self-defined private message type field .
  • Media Access Control layer Media Access Control, MAC
  • VLAN Virtual Local Area Network
  • the advanced logic processing chip TypeA will share the message Information encapsulation, the usual encapsulation format is "private message header + private message information", among which, the private message header can be realized by marking in MAC or VLAN, that is, “MAC marking + private message information” Or “MAC+VLAN marking+private message information”, you can also use the method of customizing the private message type, that is, "MAC+Private EthType (private type) + private message information", it should be noted that, for all The encapsulation of the above-mentioned message can facilitate the switch chip to identify the private message and forward it.
  • the advanced logic processing chip TypeA forwards the private message to the switching chip in the same service line card.
  • the advanced logic processing chip TypeA of the service line card 1 encapsulates the private message that needs to be shared with the advanced logic processing chip TypeB through the method of "MAC marking + private message information", and sends it to the switch of the service line card 1. chip.
  • Step 130 When the switching chip of the first service line card recognizes that the message type of the private message is a private forwarding message, determine a preset selection algorithm matching the message type of the private forwarding message, Selecting a port from ports connected to the first service line card and the switching line card according to the preset selection algorithm as a target port, and forwarding the private message to the switching line card through the target port.
  • the switch chip of the first service line card is not only used to filter input original messages to obtain messages to be processed, receive messages forwarded by switch chips of other service line cards, but also to forward the first advanced logic Process the private message processed by the chip.
  • the private message is first forwarded to the switching line card, and then the multicast copying is completed on the switching line card.
  • the switch chip of the first service line card identifies the message type, and determines that the type of message is a private message type that needs to be forwarded by multicast.
  • Step 140 the switching line card receives the private message, completes multicast replication, and forwards the copied private message to the target service line card.
  • the target service line card is a different type of service line card from the first service line card, and there may be one or more.
  • the target service line cards may be multiple service line cards of different types.
  • the switching line card receives the private message, identifies the message type, determines that the type of message is a private message type that needs multicast forwarding, creates a multicast group of the target service line card, and The copied private message is forwarded to the target service line card.
  • switching line card 1 receives a private message and recognizes that the message type is a private message type that needs to be forwarded by multicast, and creates a multicast group that needs to be shared with the target service line card where the advanced logic processing chip TypeB is located. The private message is forwarded to the target service line card.
  • the switching chip of the first service line card filters the original message to obtain the message to be processed, and sends the message to be processed to the first An advanced logic processing chip performs business processing; after the first advanced logic processing chip performs business processing on the message to be processed, it determines the private message that needs to be shared, encapsulates the private message, and forwards it to the first A switch chip of a service line card; when the switch chip of the first service line card recognizes that the message type of the private message is a private forwarding message, determine that the message type of the private forwarding message matches A preset selection algorithm, selecting a port from the ports connected to the first service line card and the switch line card according to the preset selection algorithm, as a target port, and forwarding the private message to the switch line through the target port card; the switching line card receives the private message, completes multicast replication, and forwards the copied private message to a target service line card.
  • the technical solution of this embodiment is universal, not only including private messages, but also any message with multicast forwarding requirements. Implemented in this way.
  • the forwarding scheme has no requirements for the forwarding of the backplane, whether it is the private forwarding of the same switch chip or the forwarding of heterogeneous switch chips, it can meet the requirements.
  • step 130 may be:
  • the switching chip of the first service line card obtains topology information of the first service line card and the switching line card; wherein, the topology information includes all link port information between the first service line card and all switching line cards, Link status information, and private message forwarding link selection information; wherein, the status of the link is divided into two states: open (up) and closed (down), when the state of the link port is open and there is no Private packets can be forwarded only when selected by private forwarding.
  • the switching chip of the first service line card forms an alternative port sequence for all ports whose link status is open and not selected according to the link port information;
  • the switching chip of the first service line card selects a port from the device port sequence for each switching line card according to a preset selection algorithm and forwards the current private message to the forwarding port of the switching line card, through the forwarding port Forward the private message to the switch line card, and update the link port information at the same time.
  • the selection algorithm may be a random algorithm.
  • the service line card where the advanced logic processing chip TypeA is located has topology information between the switching line card and the service line card, and the topology information includes all links between the current service line card and all switching line cards Port information, including the state of the link, information on selecting the link for private packet forwarding, etc.
  • the switching chip of the service line card can filter according to the link port information, and form a candidate port sequence ⁇ port1, port2, ..., portN ⁇ with all the ports whose link status is up and not selected by private forwarding, which can be Select one of the ports according to a specific selection algorithm as the forwarding port for forwarding the current private message to the switch line card.
  • the selection algorithm can be a random algorithm, and update the link port information at the same time to prevent other messages from using this port for forwarding. , causing path conflicts.
  • service line card 1 forms a candidate port sequence ⁇ port1, port2, ..., portN ⁇ with all ports whose link status is up and not selected by private forwarding, selects portN and forwards private packets to switch line card 1, At the same time, the link port information is updated, and the portN port has been occupied by private forwarding.
  • step 130 The advantage of setting step 130 in this way is that it optimizes the path selection from the service line card to the switch line card, avoids path conflicts, and improves the efficiency of information sharing.
  • Fig. 2 is the flow chart of a kind of message cross-board multicast copying and forwarding method based on CLOS framework that the application embodiment 2 provides, and this embodiment is the modification to above-mentioned technical solution, and this embodiment can be combined with above-mentioned one or more Multiple optional solutions in the embodiment are combined.
  • Step 210 the switching chip of the first service line card filters the original message to send the message to be processed to the first Advanced logic processing chip for business processing.
  • Step 220 After the first advanced logic processing chip performs business processing on the message to be processed, determine the private message that needs to be shared, encapsulate the private message, and forward it to the first service line card switch chip.
  • Step 230 When the switching chip of the first service line card recognizes that the message type of the private message is a private forwarding message, determine a preset selection algorithm matching the message type of the private forwarding message, Selecting a port from ports connected to the first service line card and the switching line card according to the preset selection algorithm as a target port, and forwarding the private message to the switching line card through the target port.
  • Step 240 The switching line card obtains topology information of different types of service line cards and switching line cards, the topology information includes: slot information of different types of service line cards, and the number of times a switching line card reaches each different type of service line card Link port information.
  • the different types of service line cards are different types of service line cards from the first service line card. In practical applications, multiple different types of service line cards cooperate with each other to complete the processing of a certain service. In this embodiment, after the first service line card completes service processing, it needs to forward the processed message information to the The first service line card continues to process service line cards of different types.
  • the different types of service line card slot information refers to the information of the service line card installation slot, the state of the link is divided into open (up) and closed (down) two states, when the state of the link port is open And the private message can be forwarded only when it is not selected by private forwarding.
  • Step 250 the switching line card identifies the private message type, determines the target service line card that needs to be forwarded by multicast, and obtains the target slot information of the target service line card and the backplane that reaches the target slot according to the topology information Port topology information.
  • the switching line card after the switching line card receives the forwarded private message, it can identify the type of the private message through a preset rule, and determine the type of service line card that the private message needs to share according to the type of the message, The service line card of this type is determined as the target service line card, and the target slot information where the target service line card is located and the topology information of the backplane ports reaching the target slot are obtained according to the topology information.
  • Step 260 the switching line card selects a port for each target slot according to the target slot information and the backplane port topology information, creates a private message multicast group according to the port, and uses the private message multicast group to transfer the private The message is forwarded to the target service line card to complete the copying and forwarding of the private message.
  • the switching line card selects ports whose link status is open and is not selected to form an alternative port sequence, and selects a port for each service line card according to a preset selection algorithm as the current private message forwarded to The forwarding port of the target service line card, and update the link port information at the same time.
  • the preset selection algorithm may be a random algorithm.
  • the type of the private message can be identified through the rules of the switching line card, and it is determined that the private message needs to be multicast in all Type B logic chips in the system. Therefore, the switching chip of the switching line card will create a private message multicast group, and at the same time obtain the topology information of the TypeB type in the current system.
  • the topology information includes: the slot information of the business line card where the TypeB logic chip Link port information of the service line card where the TypeB logic chip is located. Select ports whose link status is UP and which are not selected to form a candidate port sequence. For example, the port sequence from switching line card 1 to service line card 2 ⁇ L2-port1, L2-port2, ...
  • L2-portN ⁇ Switch the candidate port sequence ⁇ L3-port1, L3-port2, ... L3-portN ⁇ from line card 1 to service line card 3.
  • Select a port as the forwarding port forwarded to the service line card according to a specific selection algorithm for example, select L2-portN and L3-portN as the forwarding ports to service line card 2 and service line card 3 respectively, and update the link port information at the same time , to prevent other packets from using this port for forwarding, causing path conflicts.
  • the switching line card creates a private packet multicast group to the TypeB logical chip, and then adds the filtered ports L2-portN and L3-portN to the multicast group, and forwards the private packet type to the The multicast group completes the copying of the message, and forwards the private message to the target service line card with the help of the private message multicast group to complete the forwarding of the private message.
  • the switching line card determines that the private message needs to be multicast in all TypeB and TypeC type logic chips in the system, and the switching line card selects a service line card whose port link status is UP and is not selected
  • the ports of the switch form an alternative port sequence.
  • the port sequence ⁇ L4-port1, L4-port2, ... L4-portN ⁇ of the switching line card 1 to the service line card 4 the alternatives of the switching line card 1 to the service line card 5
  • Port sequence ⁇ L5-port1, L5-port2, ... L5-portN ⁇ , for example, select L4-portN and L5-portN as forwarding ports to reach service line card 4 and service line card 5 respectively, and add the ports to the private report Text multicast group to copy and forward private messages.
  • This embodiment changes the multicast path selection from the switch line card to the service line card, avoids path conflicts, and improves the efficiency of information sharing. All can be implemented according to this method, and the forwarding scheme has no requirements for the forwarding of the backplane, whether it is the private forwarding of the same switch chip or the forwarding of heterogeneous switch chips, it can meet the requirements.
  • Figure 3 is a flow chart of a method for copying and forwarding messages across boards based on CLOS architecture provided by Embodiment 3 of the present application. A combination of multiple options.
  • Step 310 the switching chip of the first service line card filters the original message, and sends the message to be processed to the first Advanced logic processing chip for business processing.
  • Step 320 After the first advanced logic processing chip performs business processing on the message to be processed, determine the private message that needs to be shared, encapsulate the private message, and forward it to the first service line card switch chip.
  • Step 330 When the switching chip of the first service line card recognizes that the message type of the private message is a private forwarding message, determine a preset selection algorithm matching the message type of the private forwarding message, Selecting a port from ports connected to the first service line card and the switching line card according to the preset selection algorithm as a target port, and forwarding the private message to the switching line card through the target port.
  • Step 340 After the private message is forwarded to the target service line card, if there are multiple second advanced logic processing chips in the target service line card, perform processing on the private message in the target service line card Secondary multicast forwarding.
  • the multiple second advanced logic processing chips refer to multiple advanced logic chips of the same type.
  • the target service line card After receiving the private message, the target service line card establishes an internal multicast group, and performs secondary multicast of the private message Forward to the second advanced logic processing chip.
  • the switching chip of the target service line card receives the private message, and connects the connection ports between the switching chip in the target service line card and multiple second advanced logic processing chips in the target service line card according to the port link state
  • An alternative port sequence is formed for the ports in the open state and not selected by private forwarding, each second advanced logic processing chip selects a port, creates a multicast group inside the target service line card, and uses the internal multicast group to transfer the private report
  • the message is forwarded to multiple second-level logic processing chips of the target service line card to complete the secondary multicast forwarding of the private message.
  • the private message is copied through the private message multicast group of the switching line card, and after being forwarded to the respective target service line cards, the target service line cards can process the chips according to the advanced logic on the respective service line cards
  • the number is used for secondary multicast forwarding. For example, there are two TypeB service chips on the service line card 2, so the service line card 2 will create an internal multicast group, and link the connection ports between the switch chip in the service line card and multiple logic chips according to the port
  • the ports whose state is UP and are not selected by private forwarding form an alternative port sequence, and each logic chip selects a port to form a member of the internal multicast group to complete private packet replication.
  • This embodiment changes the multicast path selection from the service line card to the advanced logic processing chip, which improves the efficiency of information sharing.
  • the solution of this embodiment is universal, and any message with multicast forwarding requirements can be processed according to the This kind of method is implemented, and the forwarding scheme has no requirements for the forwarding of the backplane, whether it is the private forwarding of the same switch chip or the forwarding of heterogeneous switch chips, it can meet the requirements.
  • FIG. 4 is a structural diagram of a message cross-board multicast replication and forwarding system based on the CLOS architecture provided in Embodiment 4 of the present application.
  • the system is used to realize the above-mentioned CLOS architecture-based message cross-board multicast replication. Forwarding methods, including:
  • At least two service line cards 41, 42 and at least one switching line card 43 each include a switching chip and at least one advanced logic processing chip, the advanced logic processing chip of the at least two service line cards
  • the types of logic processing chips are different, and each switch line card is connected to all service line cards through a backplane port; for example, the service line card 41 includes a switch chip 401 and an advanced logic processing chip 412, and the service line card 42 includes a switching chip 401 and two advanced logic processing chips 422.
  • the at least two service line cards include a first service line card and a target service line card.
  • the service line card 41 is used as the first service line card
  • the business line card 42 is the target business line card for description, wherein:
  • the switching chip 401 of the service line card 41 is set to filter the original message to obtain the message to be processed, and send the message to be processed to the senior level
  • the logic processing chip 412 performs business processing
  • the advanced logic processing chip 412 is configured to determine the private message that needs to be shared after performing business processing on the message to be processed, encapsulate the private message, and forward it to the switching chip of the service line card 41 401;
  • the switching chip 401 of the service line card 41 is also set to determine the preset selection matched by the message type of the private forwarding message when it recognizes that the message type of the private message is a private forwarding message Algorithm, according to the preset selection algorithm, select a port from the ports connected between the service line card 41 and the switch line card 43 as a target port, and forward the private message to the switch line card 43 through the target port;
  • the switching line card 43 is configured to receive the private message, and complete multicast replication, and forward the copied private message to the service line card 42;
  • the service line card 42 is configured to, after the private message is forwarded to the service line card 42, in the case of multiple advanced logic processing chips 422 in the service line card 42, process the private message Secondary multicast forwarding.

Landscapes

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

Abstract

本申请公开了一种基于CLOS架构的报文跨板组播复制转发方法和系统。该方法包括：第一业务线卡的交换芯片对原始报文进行过滤，获取待处理报文，将待处理报文按照同源同宿均衡方式发送至第一业务线卡的高级逻辑处理芯片处理，确定需要共享的私有报文并进行封装，转发给交换芯片；交换芯片识别报文类型，确定该类型所匹配的算法，根据算法从交换芯片与交换线卡连接的端口中选择目标端口转发私有报文到交换线卡；交换线卡接收私有报文完成组播复制，将私有报文转发至目标业务线卡。

Description

基于CLOS架构的报文跨板组播复制转发方法和系统
本申请要求在2021年12月06日提交中国专利局、申请号为202111511363.8的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本公开涉及通信技术领域，例如涉及一种基于CLOS架构的报文跨板组播复制转发方法和系统。
背景技术
随着网络流量的急剧增长，相关的汇聚分流设备也逐步向基于CLOS架构的方向演进，CLOS架构定义了一种几何拓扑结构，主要由主控卡、业务线卡(Line-card)和交换线卡(Fabric-card)组成。CLOS架构将控制平面和数据平面彻底分离，一方面交换线卡作为系统的转发中枢，极大地提高了系统转发能力，另一方面极大程度上提高了转发平面的可靠性，避免了控制平面的故障对转发平面的影响。
在实际应用中，根据业务复杂度的不同，同一个CLOS机框中会存在多种不同类型的业务线卡或者多块相同类型的业务线卡，它们相互配合完成某项业务的处理。在此类应用场景中，原始流量按照同源同宿的均衡算法分发给某个逻辑芯片，该逻辑芯片完成相关业务后，需要与其他业务线卡进行信息共享，通常借助系统中交换芯片的组播转发功能，以私有报文的方式实现。
但是，CLOS架构下背板带宽资源承担着整个系统的流量转发，在此场景中，私有报文的组播转发需要额外占用背板带宽资源。
发明内容
本申请实施例提供一种基于CLOS架构的报文跨板组播复制转发方法、系统，以实现CLOS架构下私有报文组播转发的功能。
第一方面，本申请实施例提供了一种基于CLOS架构的报文跨板组播复制转发方法，所述方法包括：
第一业务线卡的交换芯片对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至第一业务线卡的第一高级逻辑处理芯片进行业务处理；
所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片；
所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡；
所述交换线卡接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。
第二方面，本申请实施例提供了一种基于CLOS架构的报文跨板组播复制转发系统，包括至少两个业务线卡和至少一个交换线卡，所述至少两个业务线卡均包括交换芯片和至少一个高级逻辑处理芯片，所述至少两个业务线卡的高级逻辑处理芯片类型不同，每个交换线卡与所有的业务线卡之间通过背板端口连接，所述至少两个业务线卡包括第一业务线卡和目标业务线卡；其中：
所述第一业务线卡的交换芯片，被设置为对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至第一业务线卡的第一高级逻辑处理芯片进行业务处理；
所述第一高级逻辑处理芯片，被设置为对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给第一业务线卡的交换芯片；
所述第一业务线卡的交换芯片，还被设置为识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡；
所述交换线卡，被设置为接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。
附图说明
图1a是本申请实施例一提供的一种基于CLOS架构的报文跨板组播复制转 发方法的流程图；
图1b本申请实施例一提供的一种CLOS架构下设备系统整体框图；
图2是本申请实施例二提供的一种基于CLOS架构的报文跨板组播复制转发方法的流程图；
图3是本申请实施例三提供的一种基于CLOS架构的报文跨板组播复制转发方法的流程图；
图4是本申请实施例四提供的一种基于CLOS架构的报文跨板组播复制转发系统的示意图。
具体实施方式
下面结合附图和实施例对本申请作说明，此处所描述的实施例用于解释本申请。为了便于描述，附图中仅示出了与本申请相关的部分而非全部结构。
实施例一
图1a为本申请实施例一提供的一种基于CLOS架构的报文跨板组播复制转发方法的流程图，本实施例可适用于CLOS架构下报文跨板组播复制转发情况，该方法可以由基于CLOS架构的报文跨板组播复制转发系统来执行，包括如下步骤：
步骤110、第一业务线卡的交换芯片对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理。
在一实施方式中，图1b为CLOS架构下设备系统整体框图，系统所有的控制功能均由主控卡负责，业务线卡之间没有连线，所有的业务线卡之间的转发均通过交换线卡来实现，每个交换线卡与所有的业务线卡之间通过背板端口连接，业务线卡专注业务的处理，交换线卡则专注于系统内流量的转发。
本实施例中，所述业务线卡包括多种不同类型的业务线卡或者多块相同类型的业务线卡，所述业务线卡可以采用交换芯片和高级逻辑处理芯片的方式实现，业务线卡中的交换芯片与逻辑芯片之间的接口为多个物理链路组成的聚合口，用于流量的冗余分担，所有的交换线卡组成背板转发矩阵，每个业务线卡和交换线卡之间背板口为多个物理端口组成的背板聚合口，用于流量的冗余分担。所述交换芯片负责传统的汇聚、分流等基础功能，所述高级逻辑处理芯片实现复杂报文的深度解析等高级功能，所述高级逻辑处理芯片指可通过软件实现流量的分析、处理和转发功能的器件，例如，网络处理器(Network Processor，NP)、现场可编程逻辑门阵列(Field Programmable Gate Array，FPGA)、MIPS 多核处理器或者中央处理器(central processing unit，CPU)等，也可以是其他能够通过软件实现流量分析、处理和转发的器件。所述同源同宿均衡方式指将分布在不同链路上的同一会话的报文数据分配到同一进程处理。
可选的，所述交换芯片，基于所述交换芯片中预先设置的过滤规则对所述原始报文进行过滤，以获取待处理报文；其中，所述过滤规则可以是访问控制列表(Access Control Lists,ACL)规则。每个交换线卡与所有的业务线卡之间通过背板端口连接，故所述交换芯片通过与第一高级逻辑处理芯片之间的连接聚合端口组，将所述待处理报文上送给所述第一高级逻辑处理芯片。
可选的，所述业务面板具有面板输入端口，交换芯片对原始报文进行过滤之前，所述交换芯片通过所述第一业务线卡的面板端口获取输入的原始报文。
在一实施例中，原始报文通过所述第一业务线卡的面板端口进入到系统中，报文进入到业务线卡中的交换芯片，原始报文通过提前在交换芯片下发的ACL规则，将需要高级逻辑处理芯片TypeA处理的待处理报文过滤出来，应当说明的是，所有业务线卡的交换芯片均可以对原始报文进行过滤获得需要高级逻辑处理芯片TypeA处理的待处理报文。所述待处理报文按照同源同宿的均衡算法分发给高级逻辑处理芯片TypeA，示例性的，将所述待处理报文通过交换芯片与高级逻辑处理芯片TypeA之间的连接聚合端口组上送给高级逻辑处理芯片TypeA所在的业务线卡。示例的，原始报文进入到业务线卡1的交换芯片进行过滤获得待处理报文，待处理报文按照同源同宿的均衡方式进入业务线卡1的高级逻辑处理芯片TypeA进行处理。
步骤120、所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片。
可选的，所述第一高级逻辑处理芯片采用设定封装格式对所述私有报文进行封装，所述设定封装格式包括私有报文头和私有报文信息，所述私有报文头中的介质访问控制层(Media Access Control，MAC)字段或者虚拟局域网(Virtual Local Area Network，VLAN)字段设置有私有报文标记，或者，所述私有报文头设置有自定义的私有报文类型字段。
在一实施例中，原始报文通过高级逻辑处理芯片TypeA完成高级业务处理后，获取到的私有报文需要共享给所有的高级逻辑处理芯片TypeB，因此，高级逻辑处理芯片TypeA将共享的报文信息封装，通常的封装格式为“私有报文头+私有报文信息”，其中，私有报文头可以采用在MAC或者VLAN中打标的方式实现，即“MAC打标+私有报文信息”或者“MAC+VLAN打标+私有报文信息”，也可以采用自定义私有报文类型的方式，即“MAC+Private EthType(私 有类型)+私有报文信息”，应当说明的是，对所述报文的封装能够方便交换芯片识别该私有报文并进行转发即可。在封装完成后，高级逻辑处理芯片TypeA转发私有报文给与其处于同一业务线卡的交换芯片。示例的，业务线卡1的高级逻辑处理芯片TypeA通过“MAC打标+私有报文信息”的方式对需要共享给高级逻辑处理芯片TypeB的私有报文进行封装，发送至业务线卡1的交换芯片。
步骤130、所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡。
本实施例中，所述第一业务线卡的交换芯片不仅用于过滤输入原始报文获得待处理报文，接收其他业务线卡的交换芯片转发的报文，还用于转发第一高级逻辑处理芯片处理后的私有报文。为了避免直接在第一业务线卡上复制私有报文增加业务线卡与交换线卡的背板压力，先将私有报文转发到交换线卡上，然后在交换线卡上完成组播复制，所述第一业务线卡的交换芯片在接收到私有报文后，识别报文类型，确定该类型报文是需要组播转发的私有报文类型。
步骤140、所述交换线卡接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。
本实施例中，所述目标业务线卡是与第一业务线卡不同类型的业务线卡，可以是一个或多个。当目标业务线卡是多个时，所述目标业务线卡可以是多个不同类型的业务线卡。
在一实施例中，所述交换线卡接收所述私有报文，识别报文类型，确定该类型报文是需要组播转发的私有报文类型，创建目标业务线卡的组播组，将复制后的所述私有报文转发至目标业务线卡。示例的，交换线卡1接收私有报文并识别报文类型是需要组播转发的私有报文类型，创建需要共享给高级逻辑处理芯片TypeB所在的目标业务线卡的组播组，将所述私有报文转发到所述目标业务线卡。
本实施例中，第一业务线卡的交换芯片对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至第一业务线卡的第一高级逻辑处理芯片进行业务处理；所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给第一业务线卡的交换芯片；所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转 发到交换线卡；所述交换线卡接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。通过将私有报文转发到交换线卡上，然后在交换线卡上完成组播复制，本实施例的技术方案具有通用性，不仅包括私有报文，任何具有组播转发需求的报文均可以按照此类方法实现。同时该转发方案对背板的转发没有要求，无论是同一款交换芯片的私有转发还是异构交换芯片的转发，均可以满足要求。
在一实施例中，步骤130可选为：
所述第一业务线卡的交换芯片获取第一业务线卡与交换线卡的拓扑信息；其中，所述拓扑信息包含第一业务线卡与所有交换线卡之间的所有链路端口信息、链路的状态信息、以及私有报文转发链路选中信息；其中，所述链路的状态分为开启(up)和关闭(down)两种状态，当所述链路端口状态为开启且没有被私有转发选中时才能够进行私有报文的转发。
所述第一业务线卡的交换芯片根据链路端口信息，将所有链路状态为开启状态且没有被选中的端口组成备选端口序列；
所述第一业务线卡的交换芯片按照预设选择算法为每个交换线卡从所述设备端口序列中选择一个端口作为当前私有报文转发到交换线卡的转发端口，通过所述转发端口向交换线卡转发私有报文，同时更新链路端口信息。其中，所述选择算法可以是random算法。
在一个实施例中，高级逻辑处理芯片TypeA所在的业务线卡中有交换线卡与该业务线卡的拓扑信息，该拓扑信息中包含当前业务线卡与所有交换线卡之间的所有链路端口信息，包括链路的状态、私有报文转发链路选中信息等内容。该业务线卡的交换芯片可以根据所述链路端口信息进行筛选，将所有链路状态为up且没有被私有转发选中的端口组成一个备选端口序列{port1，port2，…,portN},可以按照某种特定的选择算法选择其中的一个端口作为当前私有报文转发到交换线卡的转发端口，所述选择算法可以是random算法，同时更新链路端口信息，防止其他报文转发使用此端口，造成路径冲突。示例的，业务线卡1将所有链路状态为up且没有被私有转发选中的端口组成备选端口序列{port1，port2，…,portN}，选择portN将私有报文转发到了交换线卡1，同时更新链路端口信息，portN端口已被私有转发占用。
步骤130这样设置的好处在于优化了业务线卡到交换线卡的路径选择，避免了路径冲突，提高了信息共享的效率。
实施例二
图2为本申请实施例二提供的一种基于CLOS架构的报文跨板组播复制转发方法的流程图，本实施例是对上述技术方案的改动，本实施例可以与上述一个或者多个实施例中的多个可选方案结合。
步骤210、第一业务线卡的交换芯片对原始报文进行过滤，以待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理。
步骤220、所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片。
步骤230、所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡。
步骤240、所述交换线卡获取不同类型业务线卡与交换线卡的拓扑信息，所述拓扑信息包括：不同类型业务线卡槽位信息、以及交换线卡到达每个不同类型业务线卡的链路端口信息。
本实施例中，所述不同类型业务线卡是与第一业务线卡不同类型的业务线卡。在实际应用中，多种不同类型的业务线卡相互配合完成某项业务的处理，在本实施例中，第一业务线卡在完成业务处理后，需要将处理后的报文信息转发给与第一业务线卡不同类型的业务线卡继续处理。所述不同类型业务线卡槽位信息指业务线卡安装槽位的信息，所述链路的状态分为开启(up)和关闭(down)两种状态，当所述链路端口状态为开启且没有被私有转发选中时才能够进行私有报文的转发。
步骤250、所述交换线卡识别私有报文类型，确定需要组播转发的目标业务线卡，并根据拓扑信息获得所述目标业务线卡所在的目标槽位信息以及到达目标槽位的背板端口拓扑信息。
在一实施例中，所述交换线卡接收转发的私有报文后，可以通过预设的规则识别私有报文的类型，根据报文的类型确定该私有报文需要共享的类型业务线卡，将该类型的业务线卡确定为目标业务线卡，根据拓扑信息获取目标业务线卡所在的目标槽位信息以及到达目标槽位的背板端口拓扑信息。
步骤260、所述交换线卡根据目标槽位信息和背板端口拓扑信息为每个目标槽位选择一个端口，根据所述端口创建私有报文组播组，借助私有报文组播组将私有报文转发到目标业务线卡，完成私有报文的复制和转发。
可选的，所述交换线卡选择端口链路状态为开启状态且没有被选中的端口组成备选端口序列，按照预设选择算法为每个业务线卡选择一个端口作为当前私有报文转发到所述目标业务线卡的转发端口，同时更新链路端口信息。所述预设选择算法可以为random算法。
在一实施例中，私有报文到达交换线卡后，可以通过交换线卡的规则识别私有报文的类型，确定该私有报文需要在系统中所有的TypeB类型的逻辑芯片中组播。因此，交换线卡的交换芯片会创建私有报文组播组，同时获得当前系统中TypeB类型的拓扑信息，该拓扑信息包括：TypeB类型逻辑芯片所在业务线卡槽位信息、交换线卡到达各个TypeB类型逻辑芯片所在业务线卡的链路端口信息。选择端口链路状态为UP且没有被选中的端口组成一个备选端口序列，示例的，交换线卡1到业务线卡2的端口序列{L2-port1，L2-port2，…L2-portN},交换线卡1到业务线卡3的备选端口序列{L3-port1，L3-port2，…L3-portN}。分别按照特定的选择算法选择一个端口作为转发到业务线卡的转发端口，例如选择L2-portN和L3-portN分别作为到达业务线卡2和业务线卡3的转发端口，同时更新链路端口信息，防止其他报文转发使用此端口，造成路径冲突。此时，交换线卡创建到TypeB类型逻辑芯片的私有报文组播组，然后将筛选的出的端口L2-portN和L3-portN加入到组播组，通过匹配私有报文类型的规则转发到组播组，完成报文的复制，借助私有报文组播组将私有报文转发到目标业务线卡，完成私有报文的转发。
在一实施例中，交换线卡确定所述私有报文需要在系统中所有的TypeB和TypeC类型的逻辑芯片中组播，交换线卡选择端口链路状态为UP且没有被选中的业务线卡的端口组成备选端口序列，示例的，交换线卡1到业务线卡4的端口序列{L4-port1，L4-port2，…L4-portN},交换线卡1到业务线卡5的备选端口序列{L5-port1，L5-port2，…L5-portN}，例如选择L4-portN和L5-portN分别作为到达业务线卡4和业务线卡5的转发端口，将所述端口添加到私有报文组播组，进行私有报文的复制和转发。
本实施例改动了交换线卡到业务线卡的组播路径选择，避免了路径冲突，提高了信息共享的效率，同时本实施例的方案具有通用性，对于任何具有组播转发需求的报文均可以按照此类方法实现，并且该转发方案对背板的转发没有要求，无论是同一款交换芯片的私有转发还是异构交换芯片的转发，均可以满足要求。
实施例三
图3为本申请实施例三提供的一种基于CLOS架构的报文跨板组播复制转 发方法的流程图，本实施例是对上述实施例的改动，可以与上述一个或者多个实施例中的多个可选方案结合。
步骤310、第一业务线卡的交换芯片对原始报文进行过滤，以待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理。
步骤320、所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片。
步骤330、所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡。
步骤340、在所述私有报文被转发至目标业务线卡之后，在所述目标业务线卡中存在多个第二高级逻辑处理芯片的情况下，在目标业务线卡中对私有报文进行二次组播转发。
本实施例中，所述多个第二高级逻辑处理芯片指多个同一类型的高级逻辑芯片，目标业务线卡接收私有报文后，组建内部组播组，将私有报文进行二次组播转发到第二高级逻辑处理芯片。
可选的，所述目标业务线卡的交换芯片接收私有报文，将目标业务线卡内交换芯片与目标业务线卡内多个第二高级逻辑处理芯片之间的连接端口，按照端口链接状态为开启状态且没有被私有转发选择的端口组成备选端口序列，每个第二高级逻辑处理芯片选择一个端口，创建所述目标业务线卡内部的组播组，借助内部组播组将私有报文转发到目标业务线卡的多个第二高级逻辑处理芯片，完成私有报文的二次组播转发。
在一实施例中，私有报文通过交换线卡的私有报文组播组完成复制，转发到各自的目标业务线卡后，目标业务线卡可以根据各自的业务线卡上的高级逻辑处理芯片的数量来进行二次组播转发。示例的，业务线卡2上存在两片TypeB类型的业务芯片，所以业务线卡2会创建内部组播组，将业务线卡内交换芯片与多个逻辑芯片之间的连接端口，按照端口链接状态为UP且没有被私有转发选择的端口组成备选端口序列，每个逻辑芯片选择一个端口，组成内部组播组的成员，完成私有报文复制。
本实施例改动了业务线卡内到高级逻辑处理芯片的组播路径选择，提高了信息共享的效率，同时本实施例的方案具有通用性，对于任何具有组播转发需 求的报文均可以按照此类方法实现，并且该转发方案对背板的转发没有要求，无论是同一款交换芯片的私有转发还是异构交换芯片的转发，均可以满足要求。
实施例四
图4为本申请实施例四提供的一种基于CLOS架构的报文跨板组播复制转发系统的结构图，该系统用于实现上文所述的基于CLOS架构的报文跨板组播复制转发系方法，包括：
至少两个业务线卡41、42和至少一个交换线卡43，所述至少两个业务线卡41、42均包括交换芯片和至少一个高级逻辑处理芯片，所述至少两个业务线卡的高级逻辑处理芯片类型不同，每个交换线卡与所有的业务线卡之间通过背板端口连接；示例的，所述业务线卡41包含交换芯片401和高级逻辑处理芯片412，所述业务线卡42包含交换芯片401和两个高级逻辑处理芯片422，所述至少两个业务线卡包括第一业务线卡和目标业务线卡，本实施例中，以业务线卡41为第一业务线卡，业务线卡42为目标业务线卡进行说明，其中：
所述业务线卡41的交换芯片401，被设置为对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至业务线卡41的高级逻辑处理芯片412进行业务处理；
所述高级逻辑处理芯片412，被设置为对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给业务线卡41的交换芯片401；
所述业务线卡41的交换芯片401，还被设置为识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述业务线卡41与交换线卡43连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡43；
所述交换线卡43，被设置为接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至业务线卡42；
所述业务线卡42，被设置为在所述私有报文被转发至业务线卡42之后，在所述业务线卡42中存在多个高级逻辑处理芯片422的情况下，对私有报文进行二次组播转发。

Claims (11)

  1. 一种基于CLOS架构的报文跨板组播复制转发方法，所述方法包括：
    第一业务线卡的交换芯片对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理；
    所述第一高级逻辑处理芯片对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片；
    所述第一业务线卡的交换芯片识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡；
    所述交换线卡接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。
  2. 根据权利要求1所述的方法，其中，所述第一业务线卡的交换芯片对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理，包括：
    所述交换芯片，基于所述交换芯片中预先设置的过滤规则对所述原始报文进行过滤，以获取待处理报文；其中，所述过滤规则包括访问控制列表规则；
    所述交换芯片通过与第一高级逻辑处理芯片之间的连接聚合端口组，将所述待处理报文上送给所述第一高级逻辑处理芯片。
  3. 根据权利要求1所述的方法，在所述第一业务线卡的交换芯片对原始报文进行过滤之前，所述方法还包括：
    所述交换芯片通过所述第一业务线卡的面板端口获取输入的原始报文。
  4. 根据权利要求1所述的方法，其中，所述第一高级逻辑处理芯片对所述私有报文进行封装，包括：
    所述第一高级逻辑处理芯片采用设定封装格式对所述私有报文进行封装，所述设定封装格式包括私有报文头和私有报文信息，所述私有报文头中的介质访问控制层MAC字段或者虚拟局域网VLAN字段设置有私有报文标记，或者，所述私有报文头设置有自定义的私有报文类型字段。
  5. 根据权利要求1所述的方法，其中，所述第一业务线卡的交换芯片根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡，包括：
    所述第一业务线卡的交换芯片获取第一业务线卡与交换线卡的拓扑信息；其中，所述拓扑信息包含第一业务线卡与所有交换线卡之间的所有链路端口信息、链路的状态信息、以及私有报文转发链路选中信息；
    所述第一业务线卡的交换芯片根据链路端口信息，将所有链路状态为开启状态且没有被选中的端口组成备选端口序列；
    所述第一业务线卡的交换芯片按照预设选择算法为每个交换线卡从所述备选端口序列中选择一个端口作为当前私有报文转发到交换线卡的转发端口，通过所述转发端口向交换线卡转发私有报文，同时更新链路端口信息。
  6. 根据权利要求1所述的方法，其中，所述交换线卡接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡，包括：
    所述交换线卡获取不同类型业务线卡与交换线卡的拓扑信息，所述拓扑信息包括：不同类型业务线卡槽位信息、以及交换线卡到达多个不同类型业务线卡的链路端口信息；
    所述交换线卡识别私有报文类型，确定需要组播转发的目标业务线卡，并根据拓扑信息获得所述目标业务线卡所在的目标槽位信息以及到达目标槽位的背板端口拓扑信息；
    所述交换线卡根据目标槽位信息和背板端口拓扑信息为每个目标槽位选择一个端口，根据所述端口创建私有报文组播组，借助私有报文组播组将私有报文转发到目标业务线卡，完成私有报文的复制和转发。
  7. 根据权利要求6所述的方法，其中，所述交换线卡根据目标槽位信息和背板端口拓扑信息为每个目标槽位选择一个端口，包括：
    所述交换线卡选择端口链路状态为开启状态且没有被选中的端口组成备选端口序列，按照预设选择算法为每个业务线卡选择一个端口作为当前私有报文转发到所述目标业务线卡的转发端口，同时更新链路端口信息。
  8. 根据权利要求1或6所述的方法，在所述私有报文被转发至目标业务线卡之后，所述方法还包括：
    在所述目标业务线卡中存在多个第二高级逻辑处理芯片的情况下，在目标业务线卡中对私有报文进行二次组播转发。
  9. 根据权利要求8所述的方法，其中，所述在目标业务线卡中对私有报文进行二次组播转发，包括：
    所述目标业务线卡的交换芯片接收私有报文，将目标业务线卡内交换芯片与目标业务线卡内多个第二高级逻辑处理芯片之间的连接端口，按照端口链接 状态为开启状态且没有被私有转发选择的端口组成备选端口序列，每个第二高级逻辑处理芯片选择一个端口，创建所述目标业务线卡内部的组播组，借助内部组播组将私有报文转发到目标业务线卡的多个第二高级逻辑处理芯片，完成私有报文的二次组播转发。
  10. 一种基于CLOS架构的报文跨板组播复制转发系统，包括至少两个业务线卡和至少一个交换线卡，所述至少两个业务线卡均包括交换芯片和至少一个高级逻辑处理芯片，所述至少两个业务线卡的高级逻辑处理芯片类型不同，每个交换线卡与所有的业务线卡之间通过背板端口连接，所述至少两个业务线卡包括第一业务线卡和目标业务线卡；其中：
    所述第一业务线卡的交换芯片，被设置为对原始报文进行过滤，以获取待处理报文，将所述待处理报文按照同源同宿均衡方式，发送至所述第一业务线卡的第一高级逻辑处理芯片进行业务处理；
    所述第一高级逻辑处理芯片，被设置为对所述待处理报文进行业务处理后，确定需要共享的私有报文，对所述私有报文进行封装，并转发给所述第一业务线卡的交换芯片；
    所述第一业务线卡的交换芯片，还被设置为识别到所述私有报文的报文类型为私有转发报文时，确定所述私有转发报文的报文类型所匹配的预设选择算法，根据所述预设选择算法从所述第一业务线卡与交换线卡连接的端口中选择端口，作为目标端口，将所述私有报文通过所述目标端口转发到交换线卡；
    所述交换线卡，被设置为接收所述私有报文，并完成组播复制，将复制后的所述私有报文转发至目标业务线卡。
  11. 根据权利要求10所述的基于CLOS架构的报文跨板组播复制转发系统，其中，所述目标业务线卡，被设置为接收所述私有报文，在所述目标业务线卡中存在多个第二高级逻辑处理芯片的情况下，对所述私有报文进行二次组播转发。
PCT/CN2022/114805 2021-12-06 2022-08-25 基于clos架构的报文跨板组播复制转发方法和系统 WO2023103461A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111511363.8 2021-12-06
CN202111511363.8A CN114157533B (zh) 2021-12-06 2021-12-06 一种基于clos架构的报文跨板组播复制转发方法和系统

Publications (1)

Publication Number Publication Date
WO2023103461A1 true WO2023103461A1 (zh) 2023-06-15

Family

ID=80450655

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/114805 WO2023103461A1 (zh) 2021-12-06 2022-08-25 基于clos架构的报文跨板组播复制转发方法和系统

Country Status (2)

Country Link
CN (1) CN114157533B (zh)
WO (1) WO2023103461A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114157533B (zh) * 2021-12-06 2023-05-26 北京锐安科技有限公司 一种基于clos架构的报文跨板组播复制转发方法和系统
CN115484217B (zh) * 2022-09-06 2024-01-05 燕山大学 基于正交架构一体化的高效动态收敛机制实现方法和系统
CN115987920A (zh) * 2022-12-23 2023-04-18 南京盛科通信有限公司 一种bier组播报文的处理方法及装置
CN116668375B (zh) * 2023-07-31 2023-11-21 新华三技术有限公司 一种报文分流方法、装置、网络设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112202674A (zh) * 2020-09-27 2021-01-08 北京物芯科技有限责任公司 一种组播报文的转发处理方法、装置、设备及存储介质
CN112398755A (zh) * 2020-11-23 2021-02-23 北京锐安科技有限公司 一种流量转发方法、业务卡和系统
CN112737982A (zh) * 2020-12-23 2021-04-30 锐捷网络股份有限公司 一种转发报文的方法和clos架构交换机
CN114157533A (zh) * 2021-12-06 2022-03-08 北京锐安科技有限公司 一种基于clos架构的报文跨板组播复制转发方法和系统

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101098238B (zh) * 2007-06-29 2010-12-22 华为技术有限公司 一种数据通信系统、交换网板及方法
CN100553199C (zh) * 2007-09-21 2009-10-21 杭州华三通信技术有限公司 基于pcie交换网的组播实现方法、系统及设备
CN102347890B (zh) * 2010-08-04 2014-03-26 杭州华三通信技术有限公司 一种组播报文的处理方法和装置
CN103338152B (zh) * 2013-05-24 2017-12-08 新华三技术有限公司 一种组播报文转发方法和主控板
CN104283817B (zh) * 2013-07-03 2017-10-27 新华三技术有限公司 用于实现交换线卡与逻辑线卡互通的方法及报文转发设备
CN108471390B (zh) * 2018-03-20 2020-12-15 深圳市风云实业有限公司 业务报文跨板处理系统及业务报文的重定向方法
CN108768866B (zh) * 2018-09-06 2021-08-03 迈普通信技术股份有限公司 组播报文跨卡转发方法、装置、网络设备及可读存储介质
CN111385026B (zh) * 2018-12-29 2022-08-26 中兴通讯股份有限公司 一种olt设备虚拟化的方法及olt设备
CN113271266B (zh) * 2021-04-21 2024-03-22 锐捷网络股份有限公司 异构交换芯片的报文转发方法及设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112202674A (zh) * 2020-09-27 2021-01-08 北京物芯科技有限责任公司 一种组播报文的转发处理方法、装置、设备及存储介质
CN112398755A (zh) * 2020-11-23 2021-02-23 北京锐安科技有限公司 一种流量转发方法、业务卡和系统
CN112737982A (zh) * 2020-12-23 2021-04-30 锐捷网络股份有限公司 一种转发报文的方法和clos架构交换机
CN114157533A (zh) * 2021-12-06 2022-03-08 北京锐安科技有限公司 一种基于clos架构的报文跨板组播复制转发方法和系统

Also Published As

Publication number Publication date
CN114157533A (zh) 2022-03-08
CN114157533B (zh) 2023-05-26

Similar Documents

Publication Publication Date Title
WO2023103461A1 (zh) 基于clos架构的报文跨板组播复制转发方法和系统
US20240179022A1 (en) Replicating broadcast, unknown-unicast, and multicast traffic in overlay logical networks bridged with physical networks
JP6317851B1 (ja) 論理ルータ
EP3533189B1 (en) Rule-based network identifier mapping
US10693784B1 (en) Fibre channel over ethernet (FCoE) link aggregation group (LAG) support in data center networks
US9331936B2 (en) Switch fabric support for overlay network features
US9083645B2 (en) Systems and methods providing reverse path forwarding compliance for a multihoming virtual routing bridge
US10887119B2 (en) Multicasting within distributed control plane of a switch
CN106850381B (zh) 交换系统、交换装置、交换机及由其执行的方法
JP6445015B2 (ja) ミドルウェアおよびアプリケーションの実行のためにエンジニアド・システムにおいてデータサービスを提供するためのシステムおよび方法
US8675522B2 (en) Conveying the VLAN/L2 VSN/bridging-domain of the incoming interface (IIF) when transporting multicast traffic over a shortest path bridged (SPB) network
US10193707B2 (en) Packet transmission method and apparatus
US8989188B2 (en) Preventing leaks among private virtual local area network ports due to configuration changes in a headless mode
US8594090B2 (en) Multicasting using a multitiered distributed virtual bridge hierarchy
US20180375799A1 (en) Multicast packet handling based on control information in software-defined networking (sdn) environment
US20150110122A1 (en) Maintaining a fabric name across a distributed switch
CN104285416A (zh) 在存储区域网络中端接覆盖隧道的虚拟路由器
CN110061915B (zh) 跨多个架构交换机的虚拟链路聚合的方法和系统
US11784922B2 (en) Scalable overlay multicast routing in multi-tier edge gateways
WO2016184283A1 (zh) 一种虚拟机数据流管理方法和系统
WO2018001242A1 (zh) 一种数据报文处理方法及装置
CN108289061B (zh) 基于sdn的业务链拓扑系统
WO2022261881A1 (zh) 一种网卡管理系统、报文处理方法及设备
CN110401726B (zh) 地址解析协议报文的处理方法、装置及设备、存储介质
WO2017036384A1 (zh) 运营商边缘设备及数据转发方法

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: 22902894

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE