WO2012000338A1 - 一种单板实现主备倒换的方法及系统 - Google Patents

一种单板实现主备倒换的方法及系统 Download PDF

Info

Publication number
WO2012000338A1
WO2012000338A1 PCT/CN2011/072736 CN2011072736W WO2012000338A1 WO 2012000338 A1 WO2012000338 A1 WO 2012000338A1 CN 2011072736 W CN2011072736 W CN 2011072736W WO 2012000338 A1 WO2012000338 A1 WO 2012000338A1
Authority
WO
WIPO (PCT)
Prior art keywords
board
main
boards
application
response
Prior art date
Application number
PCT/CN2011/072736
Other languages
English (en)
French (fr)
Inventor
张旭
方瑜
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 WO2012000338A1 publication Critical patent/WO2012000338A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/40Constructional details, e.g. power supply, mechanical construction or backplane
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/55Prevention, detection or correction of errors
    • H04L49/557Error correction, e.g. fault recovery or fault tolerance

Definitions

  • the present invention relates to a protection switching technology, and more particularly to a method and system for implementing active/standby switching of a single board. Background technique
  • the common protection switching method in the industry is in the optical cross-subsystem, where the cross-board is configured with n:m configuration, where n represents the cross-board of the actual work, and m represents the cross-board for backup. If one of the cross-boards of the n-blocks is faulty, it is immediately switched to a cross-board in the m-blocking board. This ensures that the service can be recovered in the shortest time.
  • the active/standby switchover mode requires the intervention of the control board of the NPU, such as NCP and OPM, which is responsible for the management and control of the service board.
  • the system structure is complex and the control board has problems. , will likely affect the normal implementation of the switch. Summary of the invention
  • the main purpose of the present invention is to provide a method and system for implementing active/standby switching of a single board.
  • the system has a simple structure and can easily implement active/standby switching.
  • the present invention uses the following technical solutions:
  • a method for implementing active/standby switchover on a board includes: Establish an inter-board communication channel between each board and other boards.
  • the boards are selected from the boards that are used for the main board, and the boards to be used are set as the main boards.
  • the boards are mutually negotiated, and the board to be used for the main board is selected from the board for the application.
  • the boards in turn send out at least one of the following messages: a message for requesting an active message, and responding to a primary application of another board;
  • the board that applies for the main board judges the response of the other board to its main application. When it is judged that the preset condition is reached, it determines that it is the main board.
  • the method further includes: setting a priority of the boards; the board to be used is the highest priority among the boards selected for the main application.
  • the board is used as the main board for the main board.
  • the boards are mutually negotiated, and the board to be used for the main board is selected from the board for the application.
  • the clocks are synchronized with the boards, and the sending time slots are allocated to the boards.
  • Each board sends data to the allocated sending time slots, and listens to data sent by other boards in other time slots.
  • the boards send at least one of the following messages in the allocated transmission time slot: a message for requesting the main message and responding to the main application of the other board, where one board is on the other board.
  • the response of the main application includes: judging whether the priority of the application board is higher than itself, and if yes, issuing a consent response, otherwise issuing a rejection response and issuing a main application;
  • the board that is used for the main board judges the response of the other board to its main application. When it is judged that the other boards have issued a consent response, it determines that it is the main board.
  • the boards are mutually negotiated, and the step of selecting the board to be used for the main board is used for group negotiation.
  • the board is divided into a plurality of groups, and each board in the group selects the board to be used for the main board from the board for the main application by mutual negotiation between the boards, and the selected group is selected by each group. Waiting for the main board After the boards are negotiated with each other, the board to be used is selected.
  • the predetermined condition is that the number of the agreed-upon responses is a threshold value
  • the board used by the application for determining the response of the other board to the main application includes: The number of consent responses reaches the threshold of the number of consent responses, and determines that it is the main board to be used.
  • the inter-board communication channel is a network, a system bus, or a data interface between the boards.
  • the single board is an optical cross-board.
  • the present invention also provides a system for implementing active/standby switchover between boards, and an inter-board communication channel is established between each board and other boards, including: a negotiation module for each board, for mutual negotiation between boards
  • the board to be used for the main board is selected from the board for the main application, and the board to be used for the main board is set as the main unit.
  • each board is set with a priority
  • the system includes:
  • a time slot allocation module configured to clock synchronize the boards, and allocate a transmission time slot for each board, and each board sends data in the allocated transmission time slot, and listens to data sent by other boards in other time slots;
  • the negotiation module of each board sends at least one of the following messages in the allocated transmission time slot: Applying for the primary use and responding to the primary application of the other board, where one board is on the other board
  • the main application response includes: determining whether the priority of the application board is higher than itself, and if yes, issuing a consent response; otherwise, issuing a disagree response and issuing a main application; and negotiating the module for applying for the main board
  • the board responds to the response of the main application. When it is determined that the other boards have issued a consent response, it determines that it is the main board.
  • the invention establishes an inter-board communication channel between the boards, and through the mutual negotiation between the boards, the auto-negotiation mechanism between the boards is used to select the main board from the main board for the active/standby switchover.
  • This implementation does not need to control the board intervention, which avoids the failure of the control board.
  • the risk of the active/standby switchover cannot be completed, the complexity of the system is reduced, the system construction cost is saved, the system structure is relatively simple, and the entire implementation manner is simple and easy.
  • FIG. 1 is a schematic diagram of a communication time slot of the method of the present invention.
  • 3 is a processing flow of the monitoring time slot software of the present invention.
  • FIG. 4 is a collaborative process flow of the switching process of the present invention. detailed description
  • the present invention provides a method and a system for implementing active/standby switchover on a single board to complete the protection switching of the board.
  • the main features are as follows:
  • the switchover process does not require hardware intervention such as controlling the board, but is completely self-negotiated by the board.
  • the board auto-negotiation can be completely implemented by the board software.
  • the method for implementing the active/standby switchover of the board in the embodiment of the present invention includes:
  • the boards are selected from the board to be used as the main board, and the board to be used as the main board is set as the main board.
  • the inter-board communication channel is a communication layer data interface, which is used to complete inter-board communication between boards.
  • the inter-board communication channel can take the network mode, or use the system bus mode, or the data interface between the boards.
  • the network mode is mainly applicable to mutual communication between long-distance single boards.
  • the system bus mode is to set one or more system buses, and each board is connected to the system bus.
  • the bus mode is simple in structure and easy to construct. However, in general, only one board of data can be transmitted at a time on one bus. Data interface between the boards. Each board has a data interface with other boards. You can directly negotiate data with other boards or each board. Data is negotiated only with the boards that have data interfaces. The data is negotiated with other boards that do not have data interfaces.
  • the above several modes are not mutually exclusive, and two or all of the modes may be used at the same time to ensure that the inter-board channel of the board is always smooth.
  • the negotiation messages sent by the boards mainly include: the main application message and the main application response message.
  • the main application message is the message sent by the board when the application is used for the main use.
  • the primary application response message is the response of the board to the primary application of one or more other boards, including the consent response or the rejection of the response:
  • the board can send one of the two types of negotiation messages, or both of the two types of negotiation messages can be sent. For example, a board sends a main application response message, disagrees with the main application of another board, and sends out The main application, the application itself as the main use.
  • Each board can send a negotiation message in turn.
  • the board that is used for the main board will judge the response of the other board to its main application. When the judgment reaches the preset condition, it determines that it is the main board.
  • the preset condition may be to agree to the response threshold, for example, setting the threshold to 10.
  • the board that applies for the main board judges the response of other boards to its main application. When it is judged that there are other boards
  • the response of the 10 boards is to agree to the response, that is to say, the number of agreed responses reaches the threshold of the number of agreed responses, and then it is determined that it is the main board to be used, and the setting itself is the main use.
  • the active board is quickly selected, which is suitable for the case where the active/standby switchover needs to be performed as soon as possible.
  • the priority of each board can be set, and the board with the highest priority among the boards that apply for the main board is selected as the board to be used for the main board.
  • the priority can also be used as the selection principle of the consent response or the disapproval response.
  • the board judges that the priority of the application board is higher than itself, it sends a consent response, otherwise it disagrees. Answer.
  • the above-mentioned fast negotiation mode only ensures that the highest-priority board in some boards is used as the main board, but not necessarily the highest-priority board in all boards, and one type can select the highest-priority board.
  • the embodiment process can be as follows get on:
  • the clock is synchronized with each board, and the sending time slot is allocated to each board.
  • Each board sends data to the allocated sending time slot, and listens to data sent by other boards in other time slots.
  • Each board sends at least one of the following messages in the allocated transmission time slot: Applying for the primary application and responding to the primary application of the other board; wherein, the application for the primary application of the other board to the other board
  • the response includes: judging whether the priority of the application board is higher than itself, and if yes, issuing a consent response, otherwise issuing a disagreement response and issuing a main application;
  • the board that is used for the main board judges the response of the other board to its main application. When it is judged that the other boards have issued a consent response, it determines that it is the main board.
  • each board will give a consent response when the priority of the application board is higher than itself, otherwise a disagreement is sent, and when it does not agree, it applies for the main use, and when the other boards issue a consent response, It is determined that the board is the main board for the main board. Therefore, the board to be used is the highest priority board.
  • the motherboard switching of the present invention can be implemented in the form of a protection switching protocol in a single board software.
  • Each board performs its own protocol calculation and processing.
  • the system clock synchronization is first performed.
  • the clock is the internal clock of the cross subsystem, and each cross board in the cross subsystem is allocated to a fixed time.
  • the slot (the time slot to which each board is assigned is called the transmission slot of the board).
  • the board is allowed to send data out.
  • the board can only listen to the communication. Message.
  • the main steps are as follows: Step A: Each board receives data packets through the inter-board communication channel.
  • Step B The inter-board communication channel of each board sends the received data packet to the protocol processing part of the board.
  • Step C Protocol processing
  • Step D The current status of the cross board is sent in the time slot of the board, and the status is expected (for example, the application is used, that is, the main state is expected);
  • Step E When a board in a time slot does not send data normally or data shows the board If the fault occurs, the other boards recalculate the active/standby status. You can apply for the active board to wait for the primary application in its own transmission time slot.
  • Step F Each board in the cross-subsystem determines whether there is a board for the primary application. If yes, determine whether the priority of the primary board is higher than the priority of the board. If yes, then The own transmission time slot sends a consent response; otherwise, the disagree response is sent and the main application is issued. It should be noted that the board can respond to all previous main applications, or only the previous main application. This is because, in this embodiment, after a board submits the main application, if another board submits the main application, the board that submits the application must have a higher priority than the board that applied for the previous application. Therefore, the subsequent board only needs to respond to the latter main application.
  • the main application response message is combined into one, that is, the negotiation message can be divided into the main application message (the first main application is only a simple main application, and the main application submitted later is not only the main application of the board. , and also the disagreement response to the main application of the front panel) and the main application response message (including the consent response only).
  • Step G After all the boards in the cross-subsystem agree with the main application of a board, the board determines that it is the main board to be used, and sets itself as the main board.
  • the board is divided into multiple groups. Each board in a group is elected from the board to be used by the board. Use a single board. In this way, each group can be individually negotiated in a group, and the in-group negotiation of each group can be performed simultaneously. After the group is in the group to obtain the main board for each group, the boards are to be negotiated between the groups. Therefore, the speed of the active board can be negotiated.
  • the system of the embodiment of the present invention implements an active/standby switchover system, and an inter-board communication channel is established between each board and the other boards, including: a negotiation module of each board, which is used for mutual negotiation between the boards.
  • the main board to be used is selected from the board for the main application, and the board to be used is set as the main board.
  • Each board is set with a priority, and the system may include:
  • a time slot allocation module configured to clock synchronize the boards, and allocate a transmission time slot for each board, and each board sends data in the allocated transmission time slot, and listens to data sent by other boards in other time slots;
  • the negotiation module of each board sends at least one of the following messages in the allocated transmission time slot: Applying for the primary message and responding to the primary application of the other board, where one board is on the other board.
  • the response of the main application includes: judging whether the priority of the application board is higher than itself, and if yes, issuing a consent response, otherwise issuing a disagree response and issuing a main application;
  • the negotiation module of the board that is used for the main board determines the response of the other board to its main application. When it is judged that the other boards have issued a consent response, it determines that it is the main board.
  • the present invention will be further described below by way of a specific application example.
  • the specific application example is applied to the cross-subsystem, including the n-block cross-board, and the self-negotiation of the active/standby switchover is performed by using the time slot allocation method.
  • Each board in the cross-subsystem is assigned a fixed time slot. As shown in Figure 1, the first board is assigned to slot 1, and the second board is assigned to slot 2. The number board is assigned to slot n.
  • Each board can only send packets in its own assigned time slot. In other time slots, the board can only listen to the packets sent by other boards. Status, and based on the status calculation to determine whether the main application is required.
  • the processing flow of the board sending time slots on the board includes:
  • Step 200 The board waits for the time slot of the board and processes it only in the time slot of the board.
  • Step 201 Determine whether the board is in a fault state. If it is not in a fault state, jump to the step.
  • Step 202 The board is currently in a fault state, and it is also necessary to determine whether the board is a motherboard. If it is not a motherboard, then go to step 204;
  • Step 203 The board is a faulty motherboard, and the main state is released.
  • Step 204 Notify the global board that the board is faulty and has been in the non-active state (other boards need to update the global status table), and the process proceeds to step 231.
  • Step 211 The board is a normal board. Calculate whether the board needs to be applied to the global board. If the board itself is the motherboard, or if there is a motherboard inside the current system, you do not need to apply for the master, and jump directly to step 231;
  • Step 212 When the board determines that the global motherboard can be applied, it is determined whether a board is already applying for the motherboard in the global, if no jump to step 221 is applied for the motherboard;
  • Step 213 If a board is applied to the board in the global state, the priority of the board to be applied for is higher than the board. If it is higher than the jump to 221;
  • Step 214 The priority of the board is higher than the priority of the application board. You need to notify the entire board. The board applies for the main board. Go to step 231;
  • Step 221 Agree that the current application board is the main board
  • Step 231 Report the current board status.
  • the board is a standby board, determine the current status of the active board in the system. If the active board is normal, report the current status of the board.
  • the board applies for the main use. If yes, compare the status of the application board with the status of the board to determine whether the board is preferred. The level is higher, if yes, the main application is sent; if the priority is not applied, the report agrees to answer and agrees to the main application of the board.
  • Step 300 The board enters the listening section, and determines the status of the sending board in each time slot. If the time slot is monitored, the control board of the time slot is abnormal, and the process proceeds to step 302.
  • Step 301 Receive information of a board sent by the time slot.
  • Step 302 Record abnormal information of the time slot control board.
  • Step 303 If the active board is normal at this time, go to step 306;
  • Step 304 At this time, the active board is abnormal, and it is determined whether the board has been applied for the motherboard. If not, the process goes to step 306;
  • Step 305 If the board is applied for the board, the current board is the one with the highest priority.
  • Step 306 Record the global system status and return to the listening session again.
  • the board receives the following types of packets:
  • the board in this example, the board is the board that sends the packet, the same below. This situation is divided into the following two categories:
  • the board is the main board, and the board status is judged. If the board is normal, wait for the main application (you need to wait until the sending time slot of the board to issue the main application). If the board is abnormal, continue to monitor.
  • the board is applied for the main use. Determine the priority of the board. If the board has been applied for the main board, compare the priorities of the boards and record the boards with higher priority.
  • the process flow of the switching process includes:
  • the board is the main board.
  • the application board can switch the status of the board for the main use.
  • the main priority is 1>2>3>5>6, and the single board is used.
  • the priority of the board is the highest, but there is a fault.
  • the priority of the board is the second highest and normal. The switchover needs to be switched to the board No. 2.
  • the processing flow of each board is as follows:
  • the board No. 4 has failed, and the fault status is sent in the transmission slot and the active is released.
  • the application main message is sent.
  • the main board abnormality of the 4th board is detected, and the 5th board is applied for the main use. It is judged that the board works normally but the priority is lower than that of the 5th board.
  • the main board of the board On the board of the second board, the main board of the board is abnormal.
  • the board is applied for the main board.
  • the board is working normally and the priority is higher than that of the board.
  • the 2nd board applies for the main use, and it is judged that the board works abnormally and sends a consent response.
  • the board is working in the slot 5, the main board of the board is abnormal, and the board is applied for the main board.
  • the board is working normally but the board No. 2 has a high priority.
  • the main board abnormality of the 4th board is detected, and the board No. 2 is applied for the main use. It is judged that the board works normally and the priority is not higher than the board No. 2, and the consent is sent.
  • the present invention does not need to control the board, the service board, and the hardware to participate, and does not occupy the service overhead.
  • the main board is determined by the auto-negotiation between the boards, and the structure is simple. Easy to upgrade, for example, for an existing cross subsystem, you only need to upgrade the board.
  • the software can be implemented.
  • the method has strong scalability, is not limited to fixed-scale cross protection, supports n: m protection, n and m are set according to system requirements, and will not cause major changes on the system, providing for network system expansion.
  • a simple method is not limited to fixed-scale cross protection, supports n: m protection, n and m are set according to system requirements, and will not cause major changes on the system, providing for network system expansion.
  • the above is a cross-type board.
  • the method is not limited to the active/standby switchover of the cross-connect. It can also be used for other service boards and monitoring boards.
  • This method implements 1+1 protection. Switching, 1 : n protection switching, n: m protection switching, its application scenarios are numerous.

Landscapes

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

Abstract

本发明提供了一种单板实现主备倒换的方法及系统,所述方法包括:建立各单板与其他单板之间的板间通讯通道;在需要主备倒换时,各单板通过单板间的相互协商,从申请主用的单板中推选出待主用单板,将所述待主用单板设置为主用。本发明的单板主备倒换实现方式无需控制单板干预,避免了因控制单板故障而无法完成主备倒换的风险,降低了系统的复杂度,节省了系统搭建成本,系统结构相对简单,整个实现方式简便易行。

Description

一种单板实现主^^换的方法及系统 技术领域
本发明涉及保护倒换技术, 尤指一种单板实现主备倒换的方法及系统。 背景技术
随着光网络技术的高速发展, 同步数字体系(SDH, Synchronous Digital Hierarchy ), 光传送网 (OTN, Optical Transport Network )等通讯协议和波 分复用 (WDM, Wavelength Division Multiplexing )技术的逐渐成熟, 对于 通讯线路的保护要求也越来越高。 保护倒换的一般做法是: 系统中增加一 部分冗余的备份单板, 当系统中某块单板出现异常时, 其备份单板可以在 最短的时间内接替故障单板工作, 即通过主备倒换来实现业务保护。
业务保护的种类很多, 例如对于光通讯领域中的交叉类单板(光交叉 单板), 业界通用的保护倒换方式是在光交叉子系统中, 交叉单板釆用 n:m 配置, 其中, n代表实际工作的交叉单板, m代表备份用的交叉单板。 一旦 n块工作的交叉单板中的某一块交叉单板出现故障, 立刻倒换到 m块备份 单板中的某一块交叉单板上, 这样, 保证了业务在最短时间内可以恢复。 但是, 目前这种主备倒换方式需要网元内部负责管理和控制业务单板的控 制类单板如 NCP, OPM等主控单板的干预, 系统结构较为复杂, 而且, 一 旦控制单板出现问题, 将可能影响到倒换的正常实施。 发明内容
本发明的主要目的是提供一种单板实现主备倒换的方法及系统, 系统 结构简单, 能够简便地实现主备倒换。
为解决上述技术问题, 本发明釆用了如下技术方案:
一种单板实现主备倒换的方法, 包括: 分别建立各单板与其他单板之间的板间通讯通道;
在需要主备倒换时, 各单板通过单板间的相互协商, 从申请主用的单 板中选出待主用单板, 将所述待主用单板设置为主用。
在本发明所述方法的一种实施例中, 所述各单板通过单板间相互协商, 从申请主用的单板中选出待主用单板具体包括:
所述各单板轮流发出以下消息中的至少之一: 申请主用消息、 对其他 单板的主用申请进行回应的消息;
申请主用的单板对其他单板对其主用申请的回应进行判断, 当判断达 到预设条件时, 确定自身为待主用单板。
在本发明所述方法的一种实施例中, 该方法还包括: 设置所述各单板 的优先级; 所述待主用单板为选出申请主用的单板中的优先级最高的单板 作为待主用单板。
在本发明所述方法的一种实施例中, 所述各单板通过单板间相互协商, 从申请主用的单板中推选出待主用单板具体包括:
时钟同步所述各单板, 并为各个单板分配发送时隙, 各单板在所分配 的发送时隙向外发送数据, 在其他时隙监听其他单板发送的数据;
所述各单板轮流在所分配的发送时隙发出以下消息中的至少之一: 申 请主用消息、 对其他单板的主用申请进行回应的消息, 其中, 某一单板对 其他单板的主用申请的回应包括: 判断申请主用的单板优先级是否高于自 身, 如果是, 发出同意应答, 否则发出拒绝应答, 并发出主用申请;
申请主用的单板对其他单板对其主用申请的回应进行判断, 当判断其 他单板都发出同意应答, 确定自身为待主用单板。
在本发明所述方法的一种实施例中, 所述各单板通过单板间相互协商, 从申请主用的单板中推选出待主用单板的步骤釆用分组协商, 具体包括: 将所述单板分为多个分组, 一组内的各单板通过单板间相互协商而从申请 主用的单板中推选出组内待主用单板, 各分组推选出的组内待主用单板通 过单板间相互协商而推选出待主用单板。
在本发明所述方法的一种实施例中, 所述预设条件为同意应答数阔值, 所述申请主用的单板对其他单板对其主用申请的回应进行判断包括: 当判 断其中的同意应答的数量达到所述同意应答数阔值, 确定自身为待主用单 板。
在本发明所述方法的一种实施例中, 所述板间通讯通道为网络、 系统 总线或各单板间的数据接口。
在本发明所述方法的一种实施例中, 所述单板为光交叉单板。
本发明还提供了一种单板实现主备倒换的系统,各单板与其他单板之间 建立有板间通讯通道, 包括: 各单板的协商模块, 用于通过单板间的相互 协商, 从申请主用的单板中推选出待主用单板, 将所述待主用单板设置为 主用。
在本发明所述系统的一种实施例中, 各单板设置有优先级, 所述系统 包括:
时隙分配模块, 用于时钟同步各单板, 并为各个单板分配发送时隙, 各单板在所分配的发送时隙向外发送数据, 在其他时隙监听其他单板发送 的数据;
各单板的协商模块, 轮流在所分配的发送时隙发出以下消息中的至少 一者: 申请主用、 对其他单板的主用申请进行回应, 其中, 某一单板对其 他单板的主用申请的回应包括: 判断申请主用的单板优先级是否高于自身, 如是, 发出同意应答, 否则, 发出不同意应答, 并发出主用申请; 以及 申请主用的单板的协商模块对其他单板对其主用申请的回应进行判 断, 当判断其他单板都发出同意应答, 确定自身为待主用单板。
本发明通过建立单板间的板间通讯通道, 通过各个单板之间的相互协 商, 利用单板间的自协商机制, 在发生主备倒换时, 从申请主用的单板中 选出主用单板。 这种实现方式无需控制单板干预, 避免了因控制单板故障 而无法完成主备倒换的风险, 降低了系统的复杂度, 节省了系统搭建成本, 系统结构相对简单, 整个实现方式简便易行。 附图说明
图 1是本发明方法的通讯时隙示意图;
图 2是本发明发送时隙软件处理流程;
图 3是本发明监听时隙软件处理流程;
图 4是本发明倒换处理流程协作流程。 具体实施方式
下面通过具体实施方式结合附图对本发明作进一步详细说明。
本发明提供了一种单板实现主备倒换的方法及系统, 以完成单板保护 倒换的需求, 其主要特点是: 倒换过程不需要控制单板等的硬件干预, 而 完全由单板自协商来实现主备倒换, 其接近于一种纯软件方式, 单板自协 商完全可以由单板软件来完成。
本发明实施例的单板实现主备倒换的方法, 包括:
首先, 分别建立各单板与其他单板之间的板间通讯通道;
然后, 在需要主备倒换时, 各单板通过单板间的相互协商, 从申请主 用的单板中选出待主用单板, 将所述待主用单板设置为主用。
其中, 板间通讯通道, 是通讯层数据接口, 用来完成单板之间的板间 通讯。 板间通讯通道可以釆取网络方式, 也可以釆用系统总线方式, 或者 各单板间的数据接口。 网络方式, 主要适用于远距离的单板间的相互通讯; 系统总线方式则是釆用设置一条或多条系统总线, 各单板都挂接在系统总 线上, 总线方式的结构简单, 易于构建, 但是一般一条总线上在某一时刻 只能传输一块单板的数据。 各单板间的数据接口, 可以是每一单板都与其 他单板有数据接口, 可直接与其他单板进行数据协商, 也可以是每一单板 只与部分单板有数据接口, 并通过与其有数据接口的单板过渡, 间接地与 其他与其没有数据接口的单板进行数据协商。 并且, 上述几种方式并非互 斥的, 也可以同时釆用其中的两种或全部方式, 保证单板的板间通道始终 顺畅。
协商时, 各单板发出的协商消息主要包括: 主用申请消息、 主用申请 回应消息。 其中, 主用申请消息即是单板申请主用时发出的消息, 以说明 单板申请主用。 主用申请回应消息即是单板对其他的一块或多块单板的主 用申请的响应, 包括同意应答或拒绝应答: 在同意某一单板的主用申请时, 发出同意应答, 在不同意某一单板的主用申请时, 发出拒绝应答。 单板可 以发出上述两种协商消息中的其中一种, 也可以两种协商消息都发出, 例 如, 某一单板发出主用申请回应消息, 不同意另一单板的主用申请, 同时 发出主用申请, 申请自身作为主用。
各单板可以轮流发出协商消息, 申请主用的单板会对其他单板对其主 用申请的回应进行判断, 当判断达到预设条件, 确定自身为待主用单板。 这里, 预设条件可以是同意应答数阔值, 例如设置该阔值为 10。 申请主用 的单板对其他单板对其主用申请的回应进行判断, 当判断到其他单板中有
10块单板的回应是同意应答, 也就是说同意应答的数量达到了同意应答数 阔值, 则确定自身为待主用单板, 设置自身为主用。 这种方式快速地推选 出了主用单板, 适合于需要尽快进行主备倒换的情况。
进一步地, 还可以为各单板设置优先级, 并且通过单板间相互协商从 而推选出申请主用的单板中的优先级最高的单板作为待主用单板。
对于上述快速协商方式, 也可以将优先级作为同意应答或不同意应答 的选择原则, 比如: 当单板判断申请主用的单板优先级高于自身时, 则发 出同意应答, 否则发出不同意应答。 当然, 上述快速协商方式, 只保证了 部分单板中的最高优先级单板作为主用, 但不一定是全部单板中的最高优 先级单板, 一种可以推选出最高优先级单板的实施例流程可以按如下方式 进行:
时钟同步各单板, 并为各个单板分配发送时隙, 各单板在所分配的发 送时隙向外发送数据, 在其他时隙监听其他单板发送的数据;
各单板轮流在所分配的发送时隙发出以下消息中的至少之一: 申请主 用、 对其他单板的主用申请进行回应; 其中, 某一单板对其他单板的主用 申请的回应包括: 判断申请主用的单板优先级是否高于自身, 如果是, 发 出同意应答, 否则发出不同意应答, 并发出主用申请;
申请主用的单板对其他单板对其主用申请的回应进行判断, 当判断其 他单板都发出同意应答, 确定自身为待主用单板。
由于各个单板在申请主用的单板优先级高于自身时会发出同意应答, 否则发出不同意应答, 并且在不同意时, 自身申请主用, 并在其他单板都 发出同意应答时, 才确定自身为待主用单板, 因此, 该待主用单板显然是 优先级最高的单板。
本发明主板倒换可以单板软件中的保护倒换协议形式实现。 各单板各 自进行自身的协议计算与处理相关工作。 按照上述的时隙分配法进行, 首 先进行系统时钟同步, 例如对于一个交叉子系统而言, 该时钟为交叉子系 统内部时钟, 交叉子系统中的每块交叉单板被分配到一个固定的时隙 (每 一单板被分配到的时隙称为该单板的发送时隙), 在本时隙中, 单板获准向 外发送数据, 在其它时隙中, 单板只能侦听通讯报文。 其主要步骤包括: 步骤 A: 各单板通过板间通讯通道接收数据报文;
步骤 B: 各单板的板间通讯通道将接收到的数据报文发送到单板中的 协议处理部分;
步骤 C: 协议处理;
步骤 D: 在本单板时隙发送本交叉板当前状态, 并期望状态 (例如申 请主用, 即期望为主用状态);
步骤 E:当某个时隙有一块单板没有正常发送数据或者数据显示该单板 出现故障, 其它单板重新计算主备状态, 可以申请主用的单板等待在自己 的发送时隙申请主用。
步骤 F: 交叉子系统中的每块单板判断是否有单板申请主用, 如果有, 判断该申请主用的单板的优先级是否高于本单板的优先级, 如果是, 则在 自己的发送时隙发送同意应答; 否则发送不同意应答, 并且发出主用申请。 需要说明的是, 单板可以对之前所有的主用申请都做回应, 也可以仅回应 前一个主用申请。 这是由于在本实施例中, 当一块单板提出主用申请之后, 如果又有单板提出主用申请, 则后提出申请的单板必然具有比之前提出申 请的单板更高的优先级, 因而在后的单板仅需回应后一主用申请即可。 另 一方面, 由于当一块单板提出主用申请之后, 如果又有单板提出主用申请, 则意味着后一单板不同意前一单板的主用申请, 因而可以将主用申请消息 和主用申请回应消息合而为一, 即协商消息可以分为主用申请消息 (首先 提出的主用申请仅为单纯的主用申请, 之后提出的主用申请不仅是本板的 主用申请, 同时也是对前板的主用申请的不同意应答)和主用申请回应消 息 (仅包含同意应答)。
步骤 G: 当交叉子系统中所有的单板都同意某块单板的主用申请后, 该单板确定自身为待主用单板, 并将自身设置为主用。
在单板数量较多时, 可以釆用分组协商: 将单板分为多个分组, 一组 内的各单板通过单板间相互协商而从申请主用的单板中推选出组内待主用 用单板。 这样, 各个分组可以各自进行组内协商, 多个分组各自的组内协 商可以同时进行。 在组内协商得到各个分组的组内待主用单板后, 再由这 些各分组待主用单板进行组间协商, 最终协商出待主用单板。 从而可以加 快协商得到主用单板的速度。
本发明实施例的单板实现主备倒换的系统,各单板与其他单板之间建立 有板间通讯通道, 包括: 各单板的协商模块, 用于通过单板间的相互协商, 从申请主用的单板中推选出待主用单板, 将所述待主用单板设置为主用。 各单板设置有优先级, 所述系统可以包括:
时隙分配模块, 用于时钟同步各单板, 并为各个单板分配发送时隙, 各单板在所分配的发送时隙向外发送数据, 在其他时隙监听其他单板发送 的数据;
各单板的协商模块, 轮流在所分配的发送时隙发出以下消息中的至少 之一: 申请主用、 对其他单板的主用申请进行回应消息, 其中, 某一单板 对其他单板的主用申请的回应包括: 判断申请主用的单板优先级是否高于 自身, 如是, 发出同意应答, 否则发出不同意应答, 并发出主用申请; 以 及,
申请主用的单板的协商模块对其他单板对其主用申请的回应进行判 断, 当判断其他单板都发出同意应答, 确定自身为待主用单板。
下面通过一具体应用例对本发明做进一步说明。 该具体应用例应用于 交叉子系统中, 包含 n块交叉单板, 并釆用时隙分配法进行主备倒换的自 协商。
假设在该交叉子系统中, 公用一个时钟发生源作为交叉子系统的基准 时间。 交叉子系统中的每一块单板都被分配有一个固定的时隙, 如图 1 所 示, 1号单板被分配到 1号时隙, 2号单板被分配到 2号时隙, n号单板被 分配到 n号时隙。
各单板只有在自己所分配的发送时隙中可以进行报文发送, 在其它时 隙当中仅作为监听单板监听其它单板发送的报文, 监听同时, 可以记录其 它单板报文中的状态, 并根据状态计算判断是否需要进行主用申请。
如图 2所示, 单板在本板发送时隙的处理流程包括:
步骤 200: 单板等待本单板时隙, 只有在本单板时隙中才进行处理。 步骤 201 : 判断单板是否处于故障状态, 如果非故障状态, 跳转到步骤
211 ; 步骤 202: 单板当前处于故障状态, 还需要判断一下本板是否为主板, 如果非主板, 则跳转到步骤 204;
步骤 203: 本单板是故障的主板, 释放主用状态;
步骤 204: 通知全局单板, 本板故障, 已经非主用状态(其它单板需要 更新全局状态表), 跳转到步骤 231 ;
步骤 211 : 本板为正常的单板, 计算本板是否需要申请为全局主板。 如 果本板本身就是主板, 或者当前的系统内部有主板存在, 则不需要申请为 主, 直接跳转到步骤 231 ;
步骤 212: 当本板判断可以申请全局主板时, 判断全局内是否已经有单 板正在申请主板, 如果没有跳转到步骤 221申请主板;
步骤 213: 全局内有单板申请主板, 此时需要申请单板的判断优先级是 否高于本单板, 如果高于跳转到 221 ;
步骤 214: 本板的优先级要高于申请单板的优先级, 则需要通知全体, 本板申请主用单板。 跳转到步骤 231 ;
步骤 221 : 同意当前申请单板为主用单板;
步骤 231 : 上报当前单板状态。
如图 2所示, 正常情况下, 如果当前单板为主用板, 则判断是否存在 故障, 如果有, 则释放主用状态并上报故障状态。
如果本板为备用单板, 判断系统中的当前主用单板状态, 如果主用单 板正常, 则上报本板的当前状态。
如果当前主用单板异常, 则需要判断是否有单板已经提前申请主用, 如果没有, 则本板申请主用; 如果有, 则将申请板状态与本板状态比较, 判断是否本板优先级更高, 如果是, 则发送主用申请; 如果优先级没有申 请板高, 则上报同意应答, 同意该板的主用申请。
如图 3 所示, 单板在本板监听时隙 (即发送时隙之外的其他时隙) 的 处理流程包括: 步骤 300: 本板进入监听环节, 判断每个时隙的发送单板的状态, 如果 该时隙监听超时, 则认为该时隙的控制单板异常, 跳转到步骤 302;
步骤 301 : 接收到该时隙发送单板的信息;
步骤 302: 记录该时隙控制单板的异常信息;
步骤 303: 如果此时主用单板正常, 则跳转到步骤 306;
步骤 304: 此时主用单板异常, 判断是否已经有单板申请为主板了, 如 果没有, 跳转到步骤 306;
步骤 305: 如果此时已有单板申请为主用单板, 则记录当前全局申请主 用单板优先级最高的单板信息;
步骤 306: 记录全局系统状态, 再次返回监听环节。
单板处于监听情况下, 接收到的报文类型有以下几种:
1 、 正常维护性质的报文。 处理方法是继续监听。
2 、 该单板(在本例中, 该单板即指发出报文的单板, 下同)故障。 这种情况又分为如下两种:
1 )该单板为备用单板, 则记录该单板故障状态, 继续监听。
2 )该单板为主用单板, 判断本板状态, 如果本板正常, 等待申请主用 (需要等待到本板的发送时隙才能发出主用申请), 如果本板异常, 继续监 听。
3 该单板申请主用。 判断该单板的优先级, 如果之前已经有单板申请 主用, 比较两者优先级, 记录优先级较高的单板, 继续监听。
4监听超时, 认为该单板不在位, 与故障状态处理方法一致。
如图 4所示, 倒换处理流程协作流程包括:
倒换协作处理是指, 当主板出现问题, 系统中所有的状态正常的单板 都可以申请为主用单板, 但是需要经过最终系统中在位的所有单板协商后, 决定由优先级最高的单板为主用单板。 对于某一申请主用的单板, 只有其 他单板一致通过其主用申请, 该申请板才可以切换本板状态为主用。 在图 4的示例中, 交叉系统中有 6块单板, 其中 4号单板为主用单板, 此时系统中主用优先级 1>2>3>5>6号, 1号单板优先级最高但是存在故障, 2号单板优先级次高且正常, 最终倒换需要倒换到 2号单板上。每块单板处 理流程如下:
4号单板出现故障, 在发送时隙发送故障状态且释放主用。
5号单板时隙, 检测到 4号单板主用异常, 则发送申请主用报文。
6号单板时隙, 检测到 4号单板主用异常, 且 5号单板申请主用, 判断 本板工作正常但是比 5号单板优先级低, 发送同意应答。
1号单板时隙, 检测到 4号单板主用异常, 且 5号单板申请主用, 判断 本板工作异常, 发送同意应答。
2号单板时隙, 检测到 4号单板主用异常, 且 5号单板申请主用, 判断 本板工作正常且优先级比 5号单板高, 发送申请主用。
3号单板时隙,检测到 4号单板主用异常, 且 2号单板和 5号单板申请 主用, 判断本板工作正常且优先级没有 2号单板高, 发送同意应答。
4号单板时隙, 2号单板申请主用,判断本板工作异常,发送同意应答。
5号单板时隙, 检测到 4号单板主用异常, 且 2号单板申请主用, 判断 本板工作正常但没有 2号单板优先级高, 发送同意应答。
6号单板时隙, 检测到 4号单板主用异常, 且 2号单板申请主用, 判断 本板工作正常且优先级没有 2号单板高, 发送同意应答。
1号单板时隙, 检测到 4号单板主用异常, 且 5号单板申请主用, 判断 本板工作异常, 发送同意应答。
2号单板时隙,检测到本板申请主用且优先级最高, 切换当前状态为主 用状态。
本发明相对于现有技术中的其它主备倒换方法, 不需要控制单板、 业 务单板、 硬件参与, 不占用业务开销, 完全由单板间的自协商来确定主用 单板, 结构简单, 升级方便, 例如对于现有交叉子系统, 仅需要升级单板 软件即可实现。
此外, 本方法具有很强的扩展性, 不局限于固定比例的交叉保护, 支 持 n: m的保护, n和 m根据系统需求设置, 不会引起系统上的重大变更, 为在网系统扩容提供了一种简便的方法。
以上是以交叉类单板为例, 显然, 本方法并不局限于交叉板主备倒换, 对于其它的业务类单板、 监控类单板, 也都可以釆用, 本方法实现 1+1保 护倒换、 1 : n保护倒换、 n: m保护倒换, 其应用场景众多。
以上内容是结合具体的实施方式对本发明所作的进一步详细说明, 不 能认定本发明的具体实施只局限于这些说明。 对于本发明所属技术领域的 普通技术人员来说, 在不脱离本发明构思的前提下, 还可以做出若干简单 推演或替换, 都应当视为属于本发明的保护范围。

Claims

权利要求书
1、 一种单板实现主备倒换的方法, 其特征在于, 包括:
分别建立各单板与其他单板之间的板间通讯通道;
在需要主备倒换时, 各单板通过单板间的相互协商, 从申请主用的单 板中选出待主用单板, 将所述待主用单板设置为主用。
2、 如权利要求 1所述的方法, 其特征在于, 所述各单板通过单板间相 互协商, 从申请主用的单板中选出待主用单板具体包括:
所述各单板轮流发出以下消息中的至少之一: 申请主用消息、 对其他 单板的主用申请进行回应的消息;
申请主用的单板对其他单板对其主用申请的回应进行判断, 当判断达 到预设条件时, 确定自身为待主用单板。
3、 如权利要求 2所述的方法, 其特征在于, 该方法还包括: 设置所述 各单板的优先级;
所述待主用单板为选出申请主用的单板中的优先级最高的单板。
4、 如权利要求 3所述的方法, 其特征在于, 所述各单板通过单板间相 互协商, 从申请主用的单板中推选出待主用单板具体包括:
时钟同步所述各单板, 并为各个单板分配发送时隙, 各单板在所分配 的发送时隙向外发送数据, 在其他时隙监听其他单板发送的数据;
所述各单板轮流在所分配的发送时隙发出以下消息中的至少之一: 申 请主用消息、 对其他单板的主用申请进行回应的消息, 其中, 某一单板对 其他单板的主用申请的回应包括: 判断申请主用的单板优先级是否高于自 身, 如果是, 发出同意应答, 否则发出拒绝应答, 并发出主用申请;
申请主用的单板对其他单板对其主用申请的回应进行判断, 当判断其 他单板都发出同意应答, 确定自身为待主用单板。
5、 如权利要求 1-4任一所述的方法, 其特征在于, 所述各单板通过单 板间相互协商, 从申请主用的单板中推选出待主用单板的步骤釆用分组协 商, 具体包括:
将所述单板分为多个分组, 一组内的各单板通过单板间相互协商而从 申请主用的单板中推选出组内待主用单板, 各分组推选出的组内待主用单 板通过单板间相互协商而推选出待主用单板。
6、 如权利要求 2所述的方法, 其特征在于, 所述预设条件为同意应答 数阔值;
所述申请主用的单板对其他单板对其主用申请的回应进行判断包括: 当判断其中的同意应答的数量达到所述同意应答数阔值, 确定自身为待主 用单板。
7、 如权利要求 1-4任一所述的方法, 其特征在于, 所述板间通讯通道 为网络、 系统总线或各单板间的数据接口。
8、 如权利要求 1-4任一所述的方法, 其特征在于, 所述单板为光交叉 单板。
9、一种单板实现主备倒换的系统,其特征在于, 各单板与其他单板之间 建立有板间通讯通道, 包括: 各单板的协商模块, 用于通过单板间的相互 协商, 从申请主用的单板中推选出待主用单板, 将所述待主用单板设置为 主用。
10、 如权利要求 9所述的系统, 其特征在于, 各单板设置有优先级, 所述系统包括:
时隙分配模块, 用于时钟同步各单板, 并为各个单板分配发送时隙, 各单板在所分配的发送时隙向外发送数据, 在其他时隙监听其他单板发送 的数据;
各单板的协商模块, 轮流在所分配的发送时隙发出以下消息中的至少 一者: 申请主用、 对其他单板的主用申请进行回应, 其中, 某一单板对其 他单板的主用申请的回应包括: 判断申请主用的单板优先级是否高于自身, 如是, 发出同意应答, 否则, 发出不同意应答, 并发出主用申请; 以及 申请主用的单板的协商模块对其他单板对其主用申请的回应进行判 断, 当判断其他单板都发出同意应答, 确定自身为待主用单板。
PCT/CN2011/072736 2010-07-02 2011-04-13 一种单板实现主备倒换的方法及系统 WO2012000338A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2010102165034A CN101944954A (zh) 2010-07-02 2010-07-02 一种单板实现主备倒换的方法及系统
CN201010216503.4 2010-07-02

Publications (1)

Publication Number Publication Date
WO2012000338A1 true WO2012000338A1 (zh) 2012-01-05

Family

ID=43436749

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/072736 WO2012000338A1 (zh) 2010-07-02 2011-04-13 一种单板实现主备倒换的方法及系统

Country Status (2)

Country Link
CN (1) CN101944954A (zh)
WO (1) WO2012000338A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101944954A (zh) * 2010-07-02 2011-01-12 中兴通讯股份有限公司 一种单板实现主备倒换的方法及系统
CN104038359A (zh) * 2013-03-06 2014-09-10 中兴通讯股份有限公司 虚拟交换堆叠系统的管理方法及装置
CN106612519B (zh) * 2015-10-27 2019-11-22 大唐移动通信设备有限公司 一种无线网络控制器rnc设备的单板备份方法及装置
CN107547132B (zh) * 2017-05-31 2020-12-11 新华三技术有限公司 一种板卡及分布式设备
CN110880988B (zh) * 2019-11-14 2024-04-16 国家电网有限公司 网管系统升级方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1207225A (zh) * 1996-11-12 1999-02-03 株式会社东芝 通信网络系统及其再构筑方法
CN1968075A (zh) * 2006-05-23 2007-05-23 华为技术有限公司 一种分布式热备份逻辑设备及主备板设置方法
US20090245242A1 (en) * 2008-03-31 2009-10-01 International Business Machines Corporation Virtual Fibre Channel Over Ethernet Switch
CN101944954A (zh) * 2010-07-02 2011-01-12 中兴通讯股份有限公司 一种单板实现主备倒换的方法及系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1299438C (zh) * 2002-11-02 2007-02-07 华为技术有限公司 一种实现单板主备倒换的方法及其电路

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1207225A (zh) * 1996-11-12 1999-02-03 株式会社东芝 通信网络系统及其再构筑方法
CN1968075A (zh) * 2006-05-23 2007-05-23 华为技术有限公司 一种分布式热备份逻辑设备及主备板设置方法
US20090245242A1 (en) * 2008-03-31 2009-10-01 International Business Machines Corporation Virtual Fibre Channel Over Ethernet Switch
CN101944954A (zh) * 2010-07-02 2011-01-12 中兴通讯股份有限公司 一种单板实现主备倒换的方法及系统

Also Published As

Publication number Publication date
CN101944954A (zh) 2011-01-12

Similar Documents

Publication Publication Date Title
CN101594383B (zh) 一种双控制器存储系统的服务和控制器状态监控方法
CN101257405B (zh) 主从设备间双链路的实现方法
CN103246213B (zh) 冗余控制装置之间的备用同步连接
JP4527447B2 (ja) ネットワーク中継装置及びその制御方法
WO2018233645A1 (zh) 基于CANopen协议的列车网络数据传输方法、系统及其装置
JP2005521334A (ja) ネットワーク管理システム
WO2015013896A1 (zh) 一种网络控制方法及装置
WO2007071189A1 (fr) Procede et dispositif de restauration d'un reseau maille partage
WO2014176975A1 (zh) Drni中同一端内系统之间交互信息的方法和系统
WO2012000338A1 (zh) 一种单板实现主备倒换的方法及系统
CN101217353A (zh) 一种呼叫中心多点双机冗余系统的控制方法
EP3036873A1 (en) Dedicated control path architecture for stacked packet switches
CN102098201A (zh) 一种实现l2tp用户接入备份的方法及网络系统
JP2013519268A (ja) データ転送方法、データ転送装置及びデータ転送システム
JP2015133538A (ja) データ転送装置システム、ネットワークシステム及びネットワークシステムの構成を変更する方法
WO2014090120A1 (zh) 链路切换方法及系统
US9166868B2 (en) Distributed control plane for link aggregation
WO2010048875A1 (zh) 一种集群系统扩容方法、装置及集群系统
CN109981353B (zh) 一种机框式网络通信设备中的邻站冗余保护方法及系统
CN112468328A (zh) 一种基于交换式拓扑的双余度fc-ae-1553网络重构方法
RU2563142C2 (ru) Система базовых станций для железнодорожного применения и способ формирования сети базовых станций
CN105763442A (zh) 主备倒换lacp聚合链路不中断的pon系统及方法
US20130242798A1 (en) Apparatus for duplicating router in building automatic control system and controlling method thereof
JP3880482B2 (ja) 二重化ネットワーク計算機システム及び計算機システムのネットワーク二重化方法
JP4340731B2 (ja) ネットワークの障害監視処理システム及び方法

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: 11800083

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: 11800083

Country of ref document: EP

Kind code of ref document: A1