WO2011137797A1

WO2011137797A1 - Method and system for data transmission in ethernet

Info

Publication number: WO2011137797A1
Application number: PCT/CN2011/074386
Authority: WO
Inventors: 彭华; 刘光辉; 刘永合
Original assignee: 华为技术有限公司
Priority date: 2010-07-15
Filing date: 2011-05-20
Publication date: 2011-11-10
Also published as: CN101895423A

Abstract

The embodiments of the present invention provide a method and system for data transmission in an Ethernet. The method is applied to an Ethernet service processing system, wherein the Ethernet service processing system includes at least two exchange boards and one service board, and each exchange board is linked to the service board through a physical link. The method includes: when faults occur on the Ethernet data transmission in the first exchange board, a state indication signal is transmitted to the service board through the physical link from the first exchange board to indicate that faults occur on the Ethernet data transmission in the first exchange board, so that the service board stops the Ethernet data exchange via the first exchange board, and performs the Ethernet data exchange via the second exchange board. In the embodiments of the present invention, the exchange boards transmit the state indication signal carrying the working state information of the exchange boards to the service board through the physical link, which enables the service board to obtain the working state information of the exchange boards in real time, and to select other proper exchange boards for data transmission according to the working state information of each exchange board, thus reducing the influence degree to the service.

Description

Data transmission method and system in Ethernet This application claims priority to Chinese patent application filed on July 15, 2010 by the Chinese Patent Office, Application No. 201010229830.3, entitled "Data Transmission Method and System in Ethernet" The entire contents of which are incorporated herein by reference. The present invention relates to the field of communications technologies, and in particular, to a data transmission method and system in Ethernet. BACKGROUND OF THE INVENTION Ethernet is by far the most successful local area network technology, with extensive technical support and rich hardware and software resources. Ethernet technology is widely used in office automation, industrial control, and telecommunications and computing because of its good economy, interoperability, and ease of use. In the field of telecommunications and computing, Ethernet is widely used not only outside the system, but also as a bus for data transmission between network elements. It is also widely used in the system as a bus for data transmission between boards.

Ethernet TRUNK (port aggregation) is a technology that improves transmission bandwidth and reliability. TRUNK technology binds multiple physical links into one logical link (that is, one TRUNK group), which not only improves the transmission bandwidth, but also transmits data through multiple bonded physical links. When the network fails. When one or more physical links are disconnected for other reasons, the remaining physical links can also transmit data.

A fault detection method for the Ethernet link in the prior art is as follows: The link aggregation control protocol (LACP) is used for fault detection. The LACP packets are sent to each other at the two ends of the link to determine the working status of the link. The LACP packet sending period is two: a short sending period and a long sending period. The short sending period sends an LACP packet every 1 second. A long transmission period sends an LACP packet every 30 seconds. If the local end does not receive the LACP packet sent by the peer end within 3 seconds, it determines that the peer port is faulty and determines that the link between the local end and the peer end is faulty. Thus, the traffic of the above-mentioned failed link is switched to other links in the TRUNK group. The fault detection method has at least the following problems: The LACP packet is sent according to the transmission period specified by the software protocol. When a port on the A side is faulty, the B side takes 3 seconds to detect the A side. The failure of a port, and the failure of a link to detect one of the above ports. During the above-mentioned 3 seconds, the data sent by the B side to a port on the A side through the link will be lost, resulting in packet loss and a large impact on the service.

SUMMARY OF THE INVENTION Embodiments of the present invention provide a data transmission method and system in an Ethernet network to implement data transmission between service boards in an Ethernet without packet loss.

A data transmission method in an Ethernet is applied to an Ethernet service processing system, where the Ethernet service processing system includes at least two switch boards and one service board, and each of the switch boards passes the physical link and the service The boards are connected, and the method includes:

When the first switchboard Ethernet data transmission fails, the first switch board sends a status indication signal to the service board by using the physical link to indicate that the first switchboard Ethernet data transmission fails. The service board stops the Ethernet data exchange through the first switch board, and performs Ethernet data exchange through the second switch board.

A data transmission system in an Ethernet, comprising: at least two switch board units and at least one service board, each switch board being connected to each service board through a physical link;

When the Ethernet data transmission of the first switch board fails, the first switch board sends a status indication signal to the service board through the physical link to indicate that the first switchboard Ethernet data transmission fails.

After receiving the status indicator signal sent by the switch board, the service board stops the Ethernet data exchange through the first switch board and performs Ethernet data exchange through the second switch board.

According to the technical solution provided by the embodiment of the present invention, the switch board sends a status indication signal carrying the working status information of the switch board to the service board through the physical link, because the time of transmitting the signal on the physical link is short. The service board can obtain the working status information of the switch board in time, and select other suitable switch boards according to the working status information of each switch board. Data transmission reduces the impact on the business.

BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some implementations of the present invention. For example, other drawings may be obtained from those skilled in the art without any inventive labor.

1 is a process flow diagram of a data transmission method in an Ethernet according to Embodiment 1 of the present invention;

2 is a schematic diagram of an application scenario of a service board and a switch board in an Ethernet according to Embodiment 2 of the present invention;

3 is a schematic diagram of connecting each link through a backplane connector according to Embodiment 2 of the present invention;

FIG. 4 is a flowchart of processing a data transmission method in an Ethernet according to an application scenario shown in FIG. 2 according to Embodiment 2 of the present invention;

FIG. 5 is a schematic diagram of a specific hardware implementation of a switch board according to Embodiment 3 of the present invention.

In the embodiment of the present invention, the service board obtains the working status information of the switch board according to the status indication signal of the switch board that carries the information about the working status of the switch board, and saves the obtained working status information of each switch board. Then, the service board selects a corresponding switch board for data transmission according to the working status information of each switch board.

In order to facilitate the understanding of the embodiments of the present invention, the embodiments of the present invention are not to be construed as limited.

Embodiment 1

The embodiment of the present invention provides a data transmission method in an Ethernet, which is applied to an Ethernet service processing system, where the Ethernet service processing system includes at least two switch boards and one service board. Each of the switch boards is connected to each of the service boards by using a physical link. The processing flow of the data transmission method is as shown in FIG. 1 , and includes the following steps:

The first switch board sends a status indication signal to each of the service boards by using the physical link to indicate that the Ethernet data transmission of the first switch board is faulty. ;

5102. Stop each of the service boards from performing data exchange through the first switch board, and perform Ethernet data exchange on the second switch board. Embodiment 2

FIG. 2 is a schematic diagram of an application scenario of a service board and a switch board in an Ethernet network. In FIG. 2, at least two switch boards and multiple service boards are included, and two exchanges are performed in FIG. The board and two service boards are described as an example. Referring to Figure 2, each service board is connected to the first switch board and the second switch board through an Ethernet link; each switch board is also connected to each service board through an Ethernet link, and the Ethernet in each switch board The network switching module performs Ethernet data transmission between the Ethernet link and the service processing module in each service board. Meanwhile, the first switching board and the second switching board are connected by a HiGig link.

In the embodiment of the present invention, the control module in each switchboard is further connected to each service board through one or more physical links for outputting a status indication signal. The physical link implementation form here is not limited. For example, it can be connected by using a trace on a PCB (Printed Circuit Board) or by using an external connection. In the embodiment of the present invention, in order to simplify the design of the system, the wiring on the PCB is directly connected. Signals can be transmitted quickly over physical links.

When a specific connection is made, the physical link or the Ethernet link connected to the service board can be connected to the backplane through the backplane connector of each board (service board or switch board). The traces on the backplane enable the signal to be connected.

As shown in Figure 3, the port PIN1 on the switch board is first connected to the backplane connector through the trace xl (drawn in the printed circuit board) on the board, and then connected to the backplane through the backplane connector; The port port is first connected to the backplane connector through the cable χΓ on the board, and then connected to the backplane through the backplane connector. At the same time, the signal of PIN1 and ΡΙΝΓ is realized on the backplane through the corresponding cable χ2. Connected. The above ports refer to pins of a hardware circuit based on a hardware chip such as a DSP, an FPGA, or an ASIC.

It should be noted that those skilled in the art can know from common knowledge that when connected to the backplane connector, the board is also provided with a corresponding connector for connection with the backplane connector. In the embodiment of the present invention, the connector may be referred to as a "backplane connector". Therefore, the backplane connector in the embodiment of the present invention may be connected to only the connector on the backboard. Including the connectors on the switch board (or service board), the term "backplane connector" is not strictly distinguished here. Those skilled in the art can connect the switch board (or service board) to the backplane according to the term "backplane connector".

The backplane connector here may be a variety of connectors such as a European connector, a ZD connector, an AirMax connector, a gold finger connector, etc., and is not limited herein; at the same time, the backplane connector may be connected to a port signal, or Connecting multiple port signals is not limited here.

If the status of the output signal of the switch board is strong, you can provide signals to multiple service boards through the port of one switch board, for example, one port provides signals to three service boards. Otherwise, if the status of the switch board is output. The driving capability of the signal is weak. The number of service boards that can be driven by one port is also reduced. For example, a port only provides signals to one service board. In this case, if multiple service board signals are to be provided, multiple ports are required to complete.

Based on the application scenario shown in FIG. 2, the processing flow of the data transmission method in the Ethernet provided in this embodiment is as shown in FIG. 4, and includes the following processing steps:

Step 41: The service board acquires and stores the working status of the first switch board, the second switch board, the physical layer status of the associated Ethernet port, and the link layer status.

The first switch board and the second switch board are in a normal working state. The control modules in the first switch board and the second switch board output status indication signals through the ports connected to the physical link. In the embodiment of the present invention, the state indication signal carrying the pulse level indicates that the working state is normal, and the state indicating signal carrying the fixed level indicates the working state fault; or A status indication signal carrying a first fixed level (e.g., a high level) indicates that the operating state is normal, and a status indication signal carrying a first fixed level (e.g., a low level) indicates an operational status failure. The first switch board and the second switch board are in a normal working state. Therefore, the control modules in the first switch board and the second switch board carry a pulse level or low power through a port connection of the physical link. The flat status indication signal, the status indication signal is transmitted through the physical link, and can be quickly received by each service board. When the service board acquires the pulse level or the low level carried in the status indication signal, the working state of the first switch board and the second switch board is normal according to the pulse level or the low level.

The first switch board and the second switch board respectively correspond to one slot. Each slot is associated with one or more Ethernet ports. The number of ports corresponds to the number of connected service boards. If three service boards are connected, you need to 3 Ethernet ports). Each service board can also pass Ethernet OAM.

(Operation, Administration and Maintenance, Operation, Management, and Maintenance), LACP, and other software protocols detect the physical layer status and link layer status of the Ethernet ports associated with the first switch board and the second switch board.

The first switch board or the second switch board can perform normal data exchange when the working state of the first switch board or the second switch board, the physical layer status of the associated Ethernet port, and the link layer status are all in a normal state. Waiting for work.

Each service board stores the obtained first switch board, the working state of the second switch board, the physical layer status of the associated Ethernet port, and the link layer status.

Step 42: The faulty first switch board outputs a status indication signal through the physical link, and the service board updates the stored working state of the first switch board according to the status indication signal.

When the first switch board fails due to its internal clock module, device monitoring module, or service process, the working status of the first switch board becomes faulty. Therefore, the control module in the first switchboard outputs a state indicating signal carrying a fixed level or a high level through a port connected to the physical link, where the fixed level or the high level indicates that the working state of the first switch board is faulty. .

The first switch board also turns off the HiGig link between the second switch board and the second switch board for transmitting data between the first switch board and the second switch board. Then, the first switching board starts the timer operation, and after the predetermined timing duration arrives, the first switching board performs a reset operation.

Since the service boards are connected to each switch board through the physical link, the status indication signal sent by the first switch board carrying a fixed level or a high level will be quickly received by each service board. Every The service boards obtain the fault of the working state of the first switch board according to the fixed level or high level information.

Then, each service board changes the working state information of the stored first switch board from normal to failure.

Each of the service boards also detects the physical layer status and the link layer status of the Ethernet port associated with the slot corresponding to the first switch board and the second switch board according to a predetermined detection period, and first stores the first according to the detection result. The physical layer status and link layer status of the Ethernet port associated with the slot corresponding to the switch board and the second switch board are updated.

Step 43: Each service board selects a suitable switch board for data exchange according to information such as the working state of the stored switch board, the physical layer status of the associated Ethernet port, and the link layer status.

Each service board selects an appropriate switch according to the preset switching board routing principle according to information such as the stored first switching board, the working state of the second switching board, the physical layer status of the associated Ethernet port, and the link layer status. The board exchanges data.

The above-mentioned preset switching board selection principle is expressed by the following Table 1:

First switch board

Second switch board

Working status, Ethernet

Working status, Ethernet side Conclusion Port physical layer and link layer

Port physical layer and link layer status

State

The first switch board is available, normal.

The second switch board is available. The first switch board is available, and the fault is normal.

The second switch board is unavailable. The first switch board is unavailable. The fault is normal.

The second switch board is available. The first switch board is unavailable.

The second switch board is unavailable. According to the switching board selection principle shown in Table 1, when at least one of the working state of the switch board, the physical state of the Ethernet port, and the link layer state of the Ethernet port is in a fault state, the switch board is considered as It is not available. The switch board is considered to be available only when the working status of the switch board, the physical status of the Ethernet port, and the link layer status of the Ethernet port are all in the normal state.

For the switch board that is not available, for example, in the first switch board in this embodiment, each service board sets the Ethernet link to the first switch board to be unavailable, and does not connect to the Ethernet link of the first switch board. The data is sent and the data is sent to the Ethernet link of the second switchboard. However, each service board receives data transmitted from the Ethernet link to the first switch board on the receiving side to avoid packet loss on the Ethernet link that has been sent by the first switch board.

In this embodiment, the working state of the second switchboard, the physical state of the Ethernet port, and the link layer state of the Ethernet port are all normal, and the second switchboard is available. Therefore, each service board needs to send data to the first switch board, and sends the data to the second switch board through the Ethernet link between the second switch board and the second switch board, so that the second switch board shares the faulty first switch board. Traffic.

Then, after the fault is removed, the control module in the first switchboard outputs a state indication signal carrying a pulse level or a low level through a port of the first switch board through which the physical link is connected. After the service board detects that the state of the first switch board is restored, it can send data to the Ethernet link of the switch board.

The embodiment of the present invention uses the physical link to actively send a status indication signal indicating the working status information of the switch board to the service board by using the physical link, and does not need to send according to the transmission period specified by the software protocol, because the time required for transmitting the signal on the physical link is short. Therefore, the service board can obtain the working status information of the switch board in time, and select other suitable switch boards for data transmission according to the working status information of each switch board, thereby greatly reducing the impact on the service. Embodiment 3

The embodiment of the present invention further provides a data transmission system in an Ethernet network, including: at least two switch boards and at least one service board, each switch board being connected to each service board through a physical link; When the Ethernet data transmission fails, the first switch board sends a status indication signal to the service board by using the physical link to indicate that the first switch board is faulty; After receiving the status indicator signal sent by the switch board, the service board stops data exchange through the first switch board and performs data exchange through the second switch board.

Referring to FIG. 2, an embodiment of the present invention includes two switch boards and two service boards as an example. In order to distinguish two switch boards and service boards, they are respectively named as the first switch board and the second switch board. ; and the first business board, the second business board.

Specifically, each switch board includes an Ethernet data exchange module, a control module, and other modules; each service board includes a service processing module, an indication signal processing module, and other modules.

The Ethernet data exchange module in each switch board is connected to each service board through an Ethernet link for data transmission and exchange with each service board;

The control module in each switch board is connected to the backplane through the physical link, and then connected to each service board through the backplane. When the Ethernet data transmission fails, such as the Ethernet data exchange module is faulty, or connected. In the case of a fault, a status indication signal is output to the service board through the physical link to indicate that a fault has occurred.

In a practical application, the control module may output a status indication signal carrying a pulse level to indicate that the status of the switch board is normal, and outputting a status indication signal carrying a fixed level indicating that the status of the switch board is faulty; or, output A status indication signal carrying a first fixed level (e.g., a high level) indicates that the switch board is in a normal state, and a status indication signal carrying a second fixed level (e.g., a low level) indicates that the switch board status has failed.

The indication signal processing module in each service board is configured to receive a status indication signal indicating that the first switch board is faulty outputted by the first switch board, and receive a status indication signal indicating that the second switch board is in a normal state output by the second switch board Sending, to the service processing module, a control command for stopping data exchange through the first switch board and performing data exchange through the second switch board;

The service processing module in each service board is configured to stop data exchange through the first switch board and perform data exchange through the second switch board according to the control command sent by the indication signal processing module.

Each module in the switch board or service board refers to some modules required for the normal operation of the system or according to the actual business conditions, such as the power module, which is used to provide various power supplies; and the monitoring module, which is used to monitor some parameters of the system ( Such as temperature). The physical link can be implemented as follows: The ports on each switch board are connected to the backplane connectors on the switch board through the cables on the switch board, and then connected to the backplane through the backplane connectors on the switch board. The backplane is connected to the backplane connector on the service board, and is connected to the port on the service board through the routing on the service board, thereby implementing a physical link between each of the switchboard and the service board. .

The traces on the switch board and the traces on the service board include: traces drawn in the printed circuit board. The backplane connector includes a European connector, a ZD connector, an AirMax connector, or a gold finger connector.

For the specific processing procedure of each module in the embodiment of the present invention, refer to the related description in the second embodiment, and details are not described herein again.

In the embodiment of the present invention, the specific implementation of each unit in each board (switch board or service board) can be implemented by using a dedicated chip, a general-purpose CPU, or other similar hardware chip in combination with related auxiliary circuits. limited. For example, referring to FIG. 5, it is a schematic diagram of a hardware implementation of a switchboard according to an embodiment of the present invention, including:

The processor 51, where the processor can be implemented by a general-purpose processor (such as a CPU, a DSP, an FPGA, etc.), can also be implemented by an ASIC or the like, or a special processor (such as an Ethernet processing chip). Implementation; wherein, only one processor is shown here, it can be understood that the functions of different modules (such as control module, Ethernet data exchange module) can also be completed by multiple processors.

The memory 52 is, for example, DDR (Double Data Rate), SDRAM (Synchronous Dynamic Random Access Memory), flash (flash), etc., for processor and related program operation, data storage, etc. Provides storage space. If the processor has its own internal storage module, it does not need external storage.

The auxiliary circuit 53, is the circuit required for the normal operation of the system, such as the power supply circuit, filter circuit, interface circuit, etc. required for the operation of the chip.

Similarly, the service board in the embodiment of the present invention may adopt a similar structure, and the type and type of the processor chip may be adaptively adjusted according to service requirements, and details are not described herein again. The embodiment of the present invention uses the physical link to actively send a status indication signal indicating the working status information of the switch board to the service board by using the physical link, and does not need to send according to the transmission period specified by the software protocol, because the time required for transmitting the signal on the physical link is short. Therefore, the service board can obtain the working status information of the switch board in time, and select other suitable switch boards for data transmission according to the working status information of each switch board, thereby greatly reducing the impact on the service.

In summary, in the embodiment of the present invention, the switch board actively sends a status indication signal carrying the working status information of the switch board to the service board, so that the service board can obtain the working status information of the switch board in time, and according to the working status information of each switch board. Select the appropriate switch board for data transfer. The switch can perform the active/standby switchover in a timely manner. The data transmission between the service board and the service board is not lost, and the data transmission between the switch board and the service board is not lost.

A person skilled in the art can understand that all or part of the process of implementing the above embodiment method can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium. In execution, the flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, or a read-only storage memory.

(Read-Only Memory, ROM) or Random Access Memory (RAM).

The above is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or within the technical scope disclosed by the present invention. Alternatives are intended to be covered by the scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

Rights request

A data transmission method in an Ethernet, characterized in that it is applied to an Ethernet service processing system, the Ethernet service processing system includes at least two switch boards, one service board, and each of the switch boards passes physical The link is connected to the service board, and the method includes:

2. The data transmission method in the Ethernet according to claim 1, wherein the method further comprises:

When the first switch board or the second switch board is in a normal state, the first switch board or the second switch board sends a status indication signal to each of the service boards to indicate the first switch board or The second switch board is in a normal state, so that the service board performs data exchange through the first switch board or the second switch board.

The data transmission method in the Ethernet according to claim 2, wherein the method further comprises:

The state indicating signal carrying the pulse level indicates that the state of the first switch board or the second switch board is normal, and the status indication signal carrying the fixed level indicates that the state of the first switch board or the second switch board is faulty;

Or,

The state indicating signal carrying the first fixed level indicates that the first switch board or the second switch board is in a normal state, and the state indicating signal carrying the second fixed level indicates the state of the first switch board or the second switch board. malfunction.

The data transmission method in the Ethernet according to claim 1, 2 or 3, wherein each of the switch boards is connected to the service board through a physical link, and includes:

The ports on each switch board are connected to the backplane connectors on the switch board through the cables on the switch board, and then connected to the backplane through the backplane connectors on the switch board. The backplane connectors are connected to each other and connected to the ports on the service board through the cables on the service board, thereby implementing physical links between each of the switch boards and the service boards.

The data transmission method in the Ethernet according to claim 2, wherein the service board stops data exchange through the first switch board, and performs data exchange through the second switch board. , including:

The service board receives a status indication signal indicating that the first switch board Ethernet data transmission fails by the first switch board, and obtains a working status of the first switch board as a fault; the service board receives the second exchange. Obtaining, by the physical link, the fault status indication signal of the second switch board Ethernet data transmission, obtaining the working state of the second switch board as normal;

The service board sets the Ethernet link to the first switch board to be unavailable, does not send data to the Ethernet link to the first switch board, and sends an Ethernet link to the second switch board. Send data on the road.

The data transmission method in the Ethernet according to claim 5, wherein the method further comprises:

The service board continues to receive data from an Ethernet link to the first switchboard.

A data transmission system in an Ethernet, comprising: at least two switch board units and at least one service board, each switch board being connected to each service board through a physical link; when the first switch board When the Ethernet data transmission fails, the first switch board sends a status indication signal to the service board by using the physical link to indicate that the first switchboard Ethernet data transmission fails.

8. The data transmission system of the Ethernet according to claim 7, wherein the switch board comprises:

An Ethernet data exchange module is connected to each of the service boards through an Ethernet link, and is used for data transmission and exchange with each service board;

The control module is connected to the backplane through the physical link, and is connected to each service board through the backplane. When the Ethernet data transmission fails, the status indicator is output to the service board through the physical link to indicate the fault. .

9. The data transmission system in Ethernet according to claim 8, wherein: The control module is configured to output a status indication signal carrying a pulse level, indicating that the status of the switch board is normal, and outputting a status indication signal carrying a fixed level, indicating that the status of the switch board is faulty; or The status indication signal of the level indicates that the status of the switch board is normal, and the status indication signal carrying the second fixed level indicates that the status of the switch board is faulty.

The data transmission system in the Ethernet according to claim 7, 8 or 9, wherein each of the service boards includes:

The indication signal processing module is configured to receive a status indication signal indicating that the first switch board is faulty outputted by the first switch board, and receive a status indication signal output by the second switch board indicating that the second switch board is in a normal state, and send the status indication signal to the service processing module Stopping the data exchange through the first switch board and controlling the data exchange through the second switch board;

The service processing module is configured to stop data exchange by using the first switch board according to the control command sent by the indication signal processing module, and perform data exchange by using the second switch board.