WO2010069182A1 - 实现快速重路由的方法 - Google Patents

实现快速重路由的方法 Download PDF

Info

Publication number
WO2010069182A1
WO2010069182A1 PCT/CN2009/073649 CN2009073649W WO2010069182A1 WO 2010069182 A1 WO2010069182 A1 WO 2010069182A1 CN 2009073649 W CN2009073649 W CN 2009073649W WO 2010069182 A1 WO2010069182 A1 WO 2010069182A1
Authority
WO
WIPO (PCT)
Prior art keywords
frr
group
ecmp
address
chip
Prior art date
Application number
PCT/CN2009/073649
Other languages
English (en)
French (fr)
Inventor
林宁
钱霄鸿
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
Priority to US13/140,054 priority Critical patent/US20110249679A1/en
Priority to ES09832864.4T priority patent/ES2585401T3/es
Priority to EP09832864.4A priority patent/EP2378719B1/en
Publication of WO2010069182A1 publication Critical patent/WO2010069182A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/22Alternate routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/24Multipath
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/28Routing or path finding of packets in data switching networks using route fault recovery
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/50Routing or path finding of packets in data switching networks using label swapping, e.g. multi-protocol label switch [MPLS]

Definitions

  • FIELD OF THE INVENTION The invention relates to a method of implementing fast rerouting at the driver level.
  • FRR Fast Reroote fast rerouting
  • LSP Low-power supply chain
  • the signaling and data cannot be transmitted normally. At this time, the signaling and data are transferred to another path that has been established in advance to ensure that the service will not be guaranteed. Interrupted. It can be understood that FRR is a protective measure.
  • FRR is a protective measure.
  • local protection or local protection, that is, the backup LSP only protects a part of the original LSP.
  • the protection of the nodes in the LSP is called node protection.
  • the way to protect a link in the LSP is called link protection.
  • FRR switching generally requires fast switching, one: need to be completed in 50 milliseconds, on a machine with a network processor, one: use NP for processing, but on the ASIC chip, one ⁇ : no Corresponding FRR hardware entries, so the FRR function was implemented at the driver level in the past.
  • the old old failed route is deleted, and a new route is added for forwarding.
  • the technical problem to be solved by the present invention is to provide a method for FRR, which overcomes the fact that the H drive deletes the route very many times during the FRR switching, so that the switching time cannot meet the requirements.
  • a method of fast rerouting is provided.
  • the method of the fast re-routing according to the present invention includes the following steps: a.
  • the system starts the upper layer protocol management and configures the FRR route; b.
  • the upper layer sends a valid next mega address of the FRR, and assigns a sequence number to the FRR and the ECMP group.
  • the driver writes the valid next megabyte address of the FRR into the ECMP table of the chip, and establishes a software table to store the relationship between the FRR group and the ECMP group in the corresponding chip; c. route the subnet
  • the prefix address and the FRR group number notify the driver, and the driver finds the corresponding ECMP group number according to the FRR group number in the software table, and writes the subnet routing information together with the ECMP group number into the hardware; d, the FRR group
  • the next megabyte address is faulty.
  • the upper layer software notifies the index of the FRR and the new IP address of the next megabyte.
  • the driver searches for the corresponding ECMP index according to the FRR index in the software table. The new backup is next.
  • the mega IP address is updated to the next megabyte of ECMP, and the FRR switch is completed.
  • the driver writes the ECMP group into the chip by calling the SDK function of the chip.
  • the chip's ECMP group supports writing one next megabyte, then writing one megabyte; the chip's ECMP group must support two or more next megabytes, then write two or two The same next megabyte above.
  • the next hop of the subnet route does not use the IP address, and is associated with the index of the FRR group.
  • the driver activates the FRR new valid next mega IP address by calling the SDK function of the chip. Write to the chip's ECMP table.
  • the software table in the above step b is updated.
  • the solution proposed by the present invention can implement a fast FRR switching operation method on most ASIC chips supporting ECMP, and no matter how many subnet routes are used.
  • the FRR group does not cause a boosting speed of the switching speed, which improves the response speed of the system designed by the ASIC chip to the FRR switching. It increases the efficiency of the system and makes the FRR switching speed more stable without being affected by the number of subnet routes.
  • this scheme enables the ASIC chip to support the processing of the FRR. It is not necessary to use the NP chip in order to realize the function.
  • FIG. 1 is a schematic diagram of a process for establishing a subnet route using a relationship between an FRR and an ECMP group according to the method of the present invention
  • FIG. 2 is a schematic diagram of a process flow of the method of the present invention
  • FIG. 4 is a flowchart of a method for issuing a subnet route for FRR protection according to the method of the present invention
  • the present invention proposes a method for implementing the FRR function using an ECMP table provided by an ASIC chip.
  • the BCM56624 chip provided by Broadcom is used as an example, and the solution is not limited to this switch chip.
  • the FRR fast switching function can be realized at the driver level by means of the management interface of the ECMP table.
  • the basic idea of the solution is to use FRR as an ECMP group to write to the chip.
  • this FRR switch is required. There is only one action: the next megabyte of the ECMP group corresponding to this FRR is modified once, and the next hop of the ECMP group is switched from an invalid next hop to a valid next hop, instead of The implementation of the way to delete and add multiple subnet routes.
  • Network routing operates. For example: There are n subnet routes at this time. The next hop of these subnet routes is eight and B. A is a valid next hop, and B is the backup next hop.
  • the technical solution of the present invention includes the following steps: a, the system starts the upper layer protocol management and configures the FRR route; b, the upper layer protocol delivers a valid next hop IP address of the driver layer FRR, and the protocol layer is delivered and The processing action of the driver is performed in the following three steps: bl, assigning an index number of an FRR, the index cannot be the same as the index of the ECMP group assigned by the upper layer; b2, ⁇ ) taking the IP address of the FSR as the index of the FRR Effective next megabyte, and ⁇ ) win the FRR inde and valid next mega IP address notification driver; b3, drive ⁇ 1 FRR index and valid next mega IP as an ECMP group index and this ECMP group The next mega IP address is written into the chip's ECMP table, and the ECMP group's index of the ECMP group written to the chip is saved.
  • the operation of the write chip is the same as the operation of the upper layer ECMP route. It is necessary to distinguish whether the group is actually delivered by the ECMP group or a valid next hop of the FRR in the software table of the driver. Since the FRR and the real ECMP next hop are shared when processing in the chip, Because In step b1, the index of the allocated FRR and the index of the real ECMP cannot be the same; c. If there is a subnet route, the link backup of the FRR group needs to be performed, and the prefix address and the FRR group of the subnet routes are used.
  • the notification driver the driver can find the correspondence between the FRR group number and the ECMP group number in the software table according to the FRR group number provided by the upper layer, and the subnet The information is written into the hardware together with the ECMP group number. These subnet routes point to the ECMP group in the hardware. By doing this, the next hop of the subnet route that needs to be protected by the FRR group is pointed to the hardware. The next hop of the active ECMP group; d. After completing step c, these routes to the FRR group can be protected by FRR switching.
  • the upper layer software does not need to tell the driver to refer to all the subnet routes of the FRR at this time, just need to tell the index that drives the FRR, and the new backup next megabyte IP address; d2.
  • the driver finds the index of the ECMP corresponding to the index of the FRR in the software table, updates the IP address of the next backup megabyte to the next megabyte of the ECMP, and completes the switching of the FRR; steps a and b are The upper layer writes the operation of driving the FRR group and the next hop, corresponding to FIG.
  • step c is the operation of writing the subnet route to be protected by the upper layer, corresponding to FIG. 4; step d is the time when the subnet route is switched, The upper layer is for the operation of the drive, Should 5.
  • step d is the time when the subnet route is switched, The upper layer is for the operation of the drive, Should 5.
  • steps a and b The processing of steps a and b is to create an FRR group A in the upper layer, the member of group A is 123.1.1.1, then the upper layer group A writes the driver, the driver correspondingly creates an ECMP group a, and writes 123.1.1.1 into the ECMP group a.
  • step c The processing of step c is that the upper layer sends the subnet route 123.1.1.0, and tells the driver that the next hop of the subnet route is the FRR group A, and the driver finds the ECMP group a through the FRR group A, and then 3 ⁇ 4
  • the information of "subnet route 123.1.1.0 + ECMP group a" is written into the hardware together, then the hardware can know through ECMP group a, the next megabyte of 123.1.1.0 is the next mega 123.1.1.1 corresponding to ECMP group a;
  • the processing is that if the fault occurs in 123.1.1.1, it is now switched to 123.1.1.2.
  • the upper layer sends the switching information of the FRR group A, telling the driver that the next megabyte of the FRR group A is not 123.1.1.1 but Switch to the new 123.1.1.2, then, after the driver receives, directly find the ECMP group a through the FRR group A, and the next hop of the JUC tampering ECMP group a becomes 123.1.1.2, through such a change
  • the subnet route 123.1.1.0 that points to the ECMP group a is automatically switched. Next hop.
  • step d of this method can be seen, no need Then send the detailed information of each subnet route, but directly tell the driver, the change of FRR group A, # especially can indirectly repair the next megabyte of all subnet routes.
  • FIG. 2 corresponds to the steps a and b in FIG. 2, which is a flowchart of the processing function of the FRR group: Step 101: When the upper layer protocol enables the FRR function, it needs to be sent. The FRR group and the next mega address that is currently in the enabled state.
  • Step 102 Assign a group number of the FRR different from the group number of the ECMP group sent by the upper layer to prevent the FRR group number from occupying the real ECMP.
  • Step 103 driving the group number index assigned by the FRR group as the SDK function of the inde call chip of the ECMP table of the chip, and writing the ECMP group into the chip;
  • Step 104 the driver is based on the FRR group number
  • the ECMP group is created, and the valid next hop of the FRR group is written into the ECMP group; it needs to be written according to the specific conditions of different chips. If the ECMP group of the chip supports writing a next megabyte, it can be written.
  • Step 105 create a software table in the driver, store the FRR group and corresponding ECMP group in the chip Relationship between, in order to add subnet routes, find the ERP group number corresponding to the FRR group number associated with the subnet route in the hardware;
  • Step 106 if the driver writes the hardware chip table all successfully, the return succeeds, if The write chip failed and the return failed.
  • Figure 4 corresponds to the c-step in Figure 2, which is a flow chart of the sub-route routing processing using the FRR protection.
  • Step 201 When there is a sub-route route that needs to be protected by the FRR, the upper layer needs to deliver the sub-network route to the driver.
  • Step 202 the next hop of the subnet route no longer uses the IP address, but is associated with the protection of this The index of the FRR group of the subnet route;
  • Step 203 driving the index of the FRR corresponding to the FRR by using the index of the FRR issued by the upper layer 10 in the software table established by the FRR group and the ECMP group, and further Obtaining the next megabit information of the ECMP;
  • Step 204 writing the index information of the ECMP table and the related information of the subnet routing into the subnet routing table of the hardware of the chip, and the next hop of the subnet route in the hardware points to the found The ECMP group, so the subnet route that needs FRR protection is associated with the index of the ECMP group that we replace the FRR group write hardware; the prefix of the subnet route is the content of the subnet routing table to be written
  • Step 205 if the drive write hardware chip table is all successful, the return is successful, and if the write chip fails, the return fails.
  • FIG. 5 corresponds to the step d in FIG.
  • Step 301 when the FRR is switched, the upper layer sends the FRR group and the new next hop address, the next hop. It is the backup next hop used when the FRR group is switched. Since the previous next hop has expired, the backup next hop is now in effect.
  • Step 302 find the ECMP group in the software table, and get the ECMP group number.
  • Step 303 updating the ECMP group content of the chip, and writing the new effective next mega address into the ECMP group by calling the SDK function of the chip, Substituting the next megabyte of the previous failure, the previous next megabyte no longer functions, thus completing the action of switching the next megabyte in hardware;
  • step 304 updating the relationship table between the FRR and the ECMP group in the driven software table Write the new next hop into the next hop of the ECMP group, and delete the original next hop to complete synchronization between the software table and the chip hardware table.
  • Step 305 if the drive write hardware chip table is all successful, the return is successful, and if the write chip fails, the return fails.

Landscapes

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

Description

实现快速重路由的方法 技术领域 本发明涉及一种在驱动层面实现快速重路由的方法。 背景技术 FRR ( Fast Reroote ) 快速重路由, 就是一条 LSP出现问题, 不能正常 传输信令和数据, 这时候就将信令与数据转移到事先已经建立好的另外一条 路径上去, 以保证业务不会中断。 可以这样理解 FRR是一种保护措施。 在 FRR中有两种不同类型的保护方案: 一、 路径保护, 也称为端到端保护, 这种方式是通过和现在的 LSP 并 行建立额外一条 LSP, 这条 LSP只会在发生失效时使用 二、 局部保护, 或称为本地保护, 即备份 LSP只对原 LSP的一部分进 行保护。 其中, 由于保护的对象不同, 对 LSP中的结点进行保护的方式称为 结点保护, 而对 LSP中某链路进行保护的方式称为链路保护。 目前普遍使用的 FRR切换普遍需要快速的进行切换,一^:需要在 50毫 秒完成,在拥有网络处理器的机器上,一^:使用 NP来进行处理,但是在 ASIC 芯片上, 一^:没有对应的 FRR的硬件表项, 因此以往在驱动层面实现 FRR 功能主要是通过上层在进行切换的时 H 删除以前的老的失效的路由, 再添 加一个新的路由进行转发。 这样的缺点是导致 FRR切换的时候, 如果对应某 个实效的 LSP的路由非常的多, 就会出现非常多次的要求驱动删除路由, 从 而导致切换时间不能达到要求。 目 前 中 国 专 利 CN200710166107.3 、 CN200710175346.5 、 CN200710105840, 都是从协议层面上实现的 FRR的一些方法, 其缺点是没 有考虑到芯片层面的切换速度的提高的问题。 而在无法对芯片功能进行修改 的 ASIC芯片上, 必须利用芯片已有的表项功能实现 FRR功能, 而不能像网 络处理器芯片那样编码来实现。 并且现在市场上的大部分交换芯片都没有提 供单独的 FRR切换的表项 , 而大部分的交换芯片又都提供了关于 ECMP路 由的表项管理功能。 发明内容 本发明所要解决的技术问题是提供一种 FRR的方法, 克服了在 FRR切 换的时 H 驱动非常多次地删除路由, 从而导致切换时间不能达到要求。 根据本发明的一个方面 , 提供了一种快速重路由的方法。 才艮据本发明的快速重路由的方法包括如下步骤: a、 系统启动上层协议管 理和配置 FRR路由; b、 上层下发 FRR的一条有效下一兆地址, 为 FRR分 配一个与 ECMP组的序号不相同的序号, 驱动将该 FRR的有效下一兆地址 写入芯片的 ECMP 表中, 建立一个软件表, 存储 FRR 组和对应芯片中的 ECMP组之间的关系; c、 将子网路由的前缀地址和 FRR组号通知驱动, 驱 动在所述软件表中才艮据 FRR组号查找到对应的 ECMP组号, 将子网路由信 息和 ECMP组号一起写入硬件中; d、 FRR组的有效下一兆地址出现故障, 上层软件通知驱动 FRR的 index和新的备份下一兆的 IP地址, 驱动在软件 表中才艮据 FRR的 index查找到对应的 ECMP的 index, 新的备份下一兆的 IP地址更新到 ECMP的下一兆中, 完成 FRR的切换。 在上述的步骤 b中 , 驱动通过调用芯片的 SDK函数, 将 ECMP组写入 芯片中。 在上述的步骤 b中, 芯片的 ECMP组支持写入一个下一兆, 则写入一 个下一兆; 芯片的 ECMP组必须支持两个或者两个以上下一兆, 则写两个或 两个以上相同的下一兆。 在上述的步骤 c中 , 子网路由的下一跳不使用 IP地址, 关联到 FRR组 的 index 在上述的步骤 d中 , 驱动通过调用芯片的 SDK函数, 将 FRR新的有效 下一兆 IP地址写入芯片的 ECMP表中。 在上述的步骤 d中 , 更新完 ECMP表的地址后 , 更新上述步骤 b中的 软件表。 由以上的技术方案可见, 与现有实现方法相比较, 本发明提出的方案可 以在大部分支持 ECMP的 ASIC芯片上实现快速的支持 FRR切换的操作方 法, 并且不管有多少子网路由在使用这个 FRR 组 卩不会导致切换速度的增 力口, 提高了使用 ASIC芯片设计的系统对于 FRR切换的响应速度, 大大的提 高了系统的效率 , 并且使得 FRR切换速度更加稳定 , 不会受到子网路由数量 的影响。 另外使用该方案使得 ASIC芯片一样可以支持 FRR的处理, 不需要 为了实现该功能而不得不使用 NP芯片, 使用 ASIC芯片开发的设备也比使 用 NP开发设备成本降低艮多。 本发明的其它特征和优点将在随后的说明书中阐述, 并且, 部分地从说 明书中变得显而易见, 或者通过实施本发明而了解。 本发明的目的和其他优 点可通过在所写的说明书、 权利要求书、 以及附图中所特别指出的结构来实 现和获得。 附图说明 附图用来提供对本发明的进一步理解, 并且构成说明书的一部分, 与本 发明的实施例一起用于解释本发明 , 并不构成对本发明的限制。 在附图中: 图 1为本发明方法的子网路由利用 FRR与 ECMP组的关系建立切换的 示意图; 图 2为本发明方法的处理流程总图; 图 3为本发明方法的 FRR组下发处理流程图; 图 4为本发明方法的使用 FRR保护的子网路由下发处理流程图; 图 5为本发明方法的 FRR切换下一兆上层下发处理流程图。 具体实施方式 功能相克述 本发明提出了一种使用 ASIC芯片提供的 ECMP表的来实现 FRR功能 的方法, 本文中使用 Broadcom公司提供的 BCM56624芯片来进行举例, 该 方案不仅仅限于这一款交换芯片, 只要芯片厂商提供了 ECMP 的表和针对 ECMP表的管理接口 , 就可以在驱动层面上借助 ECMP表的管理接口来实现 FRR的快速切换的功能。 该方案实现的基本思路就是 FRR 作为一个 ECMP 组写入芯片的
ECMP表中, 并且无论有多少条子网路由使用了这个 FRR的切换, 所需要做 的动作仅仅只有一个: 这个 FRR对应的 ECMP组的下一兆内容进行一次 修改,把这个 ECMP组的下一跳从一个无效的下一跳切换到一个有效下一跳 即可, 而无需像以前的实现方式那样进行多次子网路由的删除和添加。 如图 1所示, 尽管有多条子网路由指向需要切换的 FRR组, 但是如果 使用 ECMP组来实现,在切换的时候仅仅需要将 ECMP组的下一跳进行切换 即可, 不需要对所有子网路由进行操作。 例如: 此时有 n条子网路由, 这些 子网路由的下一跳有八、 B两个, A是有效的下一跳, B是备份下一跳, 而 此时仅需要创建一个 ECMP组,该组的下一跳是 A,那么此时一旦开始切换, 仅仅需要在这个 ECMP组中把下一跳切换为 B, 而所有的子网路由都不需要 删除和添加就可以完成这个 FRR的动作,对应图 1仅仅需要将实线的下一跳 切换到虚线的下一兆, 即可实现 FRR。 如图 2, 本发明的技术方案包括如下步骤: a、 系统启动上层协议管理和配置 FRR路由; b、 上层协议下发驱动层 FRR的一条有效的下一跳 IP地址, 协议层的 下发以及驱动的处理动作是按以下三个步骤进行的: bl、 分配一个 FRR的序号 index, 该 index与上层分配的 ECMP组的序 号 index不能相同; b2、 ^)夺下发的 IP地址作为 FRR的 index的有效下一兆, 并^)夺该 FRR的 inde 和有效下一兆 IP地址通知驱动; b3、驱动^ 1 FRR的 index和有效下一兆 IP当作是一个 ECMP组的 index 和这个 ECMP组的下一兆 IP地址写入芯片的 ECMP表中,并保存这个 ECMP 组写入芯片的 ECMP组的 index; 此处驱动方面写芯片的操作与操作上层下 发 ECMP路由的方法是一样的 ,仅仅需要在驱动的软件表中区分这个组到底 是真正的 ECMP组下发的还是一个 FRR的有效下一跳下发的, 由于在芯片 中处理的时候 , FRR和真正的 ECMP下一跳组是共用的, 因此在步骤 b 1中, 分配的 FRR的 index和真正的 ECMP的 index不能相同; c、 此时如果有子网路由需要进行 FRR组的链路备份, 将这些子网路由 的前缀地址和 FRR组号通知驱动, 驱动根据上层提供的 FRR组号, 就可以 在所述软件表中查找到 FRR组号和 ECMP组号之间的对应关系, 将子网路 由信息和 ECMP 组号一起写入硬件中, 这些子网路由就指向了硬件中的 ECMP组,通过这样的操作也就将需要 FRR组保护的子网路由的下一跳都指 向了硬件中事实上起作用的 ECMP组的下一跳; d、 当完成步骤 c之后, 这些指向 FRR组的路由就可以进行保护的 FRR 切换了 , 此时 , 如果该 FRR组的有效的下一跳出现故障需要切换到备份下一 跳, 动作 口下: dl、上层软件不需要再告诉驱动此时引用了该 FRR的所有的子网路由 , 仅仅需要告诉驱动该 FRR的 index, 和新的备份下一兆的 IP地址; d2、 驱动在软件表中找到该 FRR的 index对应的 ECMP的 index, 新 的备份下一兆的 IP地址更新到这个 ECMP的下一兆中, 完成 FRR的切换; 步骤 a、 b是上层写入驱动 FRR组和下一跳的操作, 对应图 3; 步骤 c 是上层写入需要被保护的子网路由的操作, 对应图 4; 步骤 d是子网路由进 行切换的时 4矣 , 上层对于驱动的操作 , 对应图 5。 为了更清楚地理解本发明所述技术方案, 现举例如下: 如有一个子网路 由 123.1.1.0 , 它有两个下一兆, 正在使用的下一兆是 123.1.1.1 , 备份下一兆 是 123.1.1.2。 步骤 a、 b的处理,是上层创建一个 FRR组 A,组 A的成员是 123.1.1.1 , 然后上层 组 A写入驱动, 驱动对应的创建一个 ECMP组 a, 将 123.1.1.1 写入 ECMP组 a中; 步骤 c的处理, 是上层下发子网路由 123.1.1.0 , 并且告诉驱动这个子网 路由的下一跳是 FRR组 A, 驱动就通过这个 FRR组 A找到了 ECMP组 a, 然后 ·¾"子网路由 123.1.1.0 + ECMP组 a"的信息一起写入硬件, 则硬件可 以通过 ECMP组 a知道, 123.1.1.0 的下一兆是 ECMP组 a对应的下一兆 123.1.1.1; 步骤 d的处理, 是如果 123.1.1.1 出现故障了, 现在要切换到 123.1.1.2 , 这时, 上层下发 FRR组 A的切换信息, 告诉驱动 FRR组 A现在的下一兆已 经不是 123.1.1.1 而是切换成新的 123.1.1.2 了, 于是, 驱动收到之后, 直接 通过 FRR组 A找到 ECMP组 a,并 JU'爹改 ECMP组 a的下一跳变成 123.1.1.2 , 通过这样的^ "改,刚才指向 ECMP组 a的子网路由 123.1.1.0就自动的切换了 下一跳。 本方法的好处就是,如果指向这个 FRR组 A的路由很多,除了 123.1.1.0 , 还有艮多其他的子网路由都指向这个 FRR 组, 那么通过这种方法的步骤 d 可以看出,不需要再下发每个子网路由的详细信息, 而是直接告诉驱动, FRR 组 A的变化 , #尤可以间接的修_丈所有子网路由的下一兆。 下面就图 2的步骤作具体的描述, 图 3对应图 2中的 a、 b步骤,是 FRR 组下发处理函数流程图: 步骤 101 , 当上层协议使能了 FRR功能的时候 , 需要下发 FRR组和目 前处在使能状态的下一兆地址; 步骤 102 ,分配一个与真正的上层下发的 ECMP组的组号不一样的 FRR 的组号, 防止出现 FRR组号占用了真正的 ECMP组组号的情况; 步骤 103 , 驱动将该 FRR组分配的组号 index作为芯片的 ECMP表的 inde 调用芯片的 SDK函数, 将 ECMP组写入芯片中; 步骤 104, 驱动才艮据 FRR组号分配创建了 ECMP组, 将所述 FRR组的 有效下一跳写入该 ECMP组; 这里需要根据各个不同芯片的具体情况写入 , 如果芯片的 ECMP组支持写入一个下一兆则可以写入一个下一兆, 如果必须 支持两个或两个以上下一兆 , 则可以写两个或两个以上相同的下一兆; 步骤 105 , 在驱动中建立一个软件表, 存储这个 FRR组和对应的芯片 中的 ECMP组之间的关系, 以便子网路由添加的时候, 查找子网路由关联的 FRR的组号在硬件中应该对应的是哪个 ECMP组号; 步骤 106 , 如果驱动写硬件芯片表全部成功, 返回成功, 如果写芯片失 败, 返回失败。 图 4对应图 2中的 c步骤, 是使用 FRR保护的子网路由下发处理流程 图: 步骤 201 , 当有需要 FRR保护的子网路由出现的时候, 上层需要下发 这种子网路由给驱动; 步骤 202 , 子网路由的下一跳不再使用 IP地址, 而是关联到保护这个 子网路由的 FRR组的 index; 步骤 203 , 驱动在所述 FRR组与 ECMP组建立的软件表中, 通过上层 十办议下发的 FRR的 index, 查找该 FRR对应的 ECMP组的 index, 进而得到 ECMP的下一兆信息; 步骤 204, 将 ECMP表的 index信息和子网路由的相关信息写入芯片的 硬件的子网路由表中, 硬件中的这条子网路由的下一跳就指向该找到的 ECMP组 , 这样就 需要 FRR保护的这个子网路由和我们替代 FRR组写入 硬件的 ECMP组的 index联系起来了; 子网路由的前缀就是要写入硬件的子网路由表的内容, 例如 123.0.0.0 就是一个子网路由前缀; 子网路由表指的是硬件的子网路由表。 子网路由写 入的时候, 如果是 FRR的类型 , 需要找到一个对应的 ECMP表的 index作为 该子网路由的下一跳写入硬件 , 硬件会知道这个子网路由对应的下一跳是一 个 ECMP组, 然后就会在 ECMP表中找该 ECMP组, 该 ECMP组又记录了 1个或者多个下一兆, 这 1 个或者多个下一兆就是该子网路由的真实的下一 跳。 步骤 205 , 如果驱动写硬件芯片表全部成功, 返回成功, 如果写芯片失 败, 返回失败。 图 5对应图 2中的 d步骤, 是 FRR切换下一跳上层下发处理流程图: 步骤 301 , 当 FRR发生切换的时候, 上层下发 FRR组以及新的下一跳 地址, 这个下一跳是用于这个 FRR组切换时候的备份下一跳, 由于以前的下 一跳已经失效, 因此这个备份下一跳现在生效; 步骤 302 , 在软件表中找到这个 ECMP组, 得到 ECMP组的组号, 并 且将新的需要生效的下一跳准备写入硬件中; 步骤 303 , 更新芯片的 ECMP组内容, 通过调用芯片的 SDK函数, 把 新的生效的下一兆地址写入这个 ECMP组中, 替代以前的失效的下一兆, 以 前的下一兆不再产生作用, 这样就完成了在硬件中切换下一兆的动作; 步骤 304, 更新驱动的软件表中的 FRR与 ECMP组的关系表, 将新的 下一跳写入 ECMP组的下一跳中 , 并且删除原来的下一跳 , 完成软件表与芯 片硬件表的同步; 步骤 305 , 如果驱动写硬件芯片表全部成功, 返回成功, 如果写芯片失 败, 返回失败。 以上所述仅为本发明的优选实施例而已, 并不用于限制本发明, 对于本 领域的技术人员来说, 本发明可以有各种更改和变化。 凡在本发明的^^申和 原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明的保护 范围之内。

Claims

权 利 要 求 书
1. 一种实现快速重路由的方法, 其特征在于, 它包括如下步骤:
a、 系统启动上层十办议管理和配置 FRR路由;
b、 上层下发 FRR的一条有效下一兆地址, 为 FRR分配一个与 ECMP组的序号不相同的序号, 驱动将该 FRR的有效下一兆地址写入 芯片的 ECMP表中, 建立一个软件表, 存储 FRR组和对应芯片中的 ECMP组之间的关系;
c、 将子网路由的前缀地址和 FRR组号通知驱动, 驱动在所述软 件表中才艮据 FRR组号查找到对应的 ECMP组号, 将子网路由信息和 ECMP组号一起写入硬件中;
d、 FRR组的有效下一兆地址出现故障, 上层软件通知驱动 FRR 的 index和新的备份下一跳的 IP地址, 驱动在所述软件表中才艮据 FRR 的 index查找到对应的 ECMP的 index, 新的备份下一兆的 IP地址 更新到 ECMP的下一兆中, 完成 FRR的切换。
2. 按照权利要求 1所述的实现快速重路由的方法, 其特征在于: 在所述 的步骤 b中 , 驱动通过调用芯片的 SDK函数, 将 ECMP组写入芯片 中。
3. 按照权利要求 1所述的实现快速重路由的方法, 其特征在于: 在所述 的步骤 b中, 芯片的 ECMP组支持写入一个下一跳, 则写入一个下一 跳; 芯片的 ECMP组必须支持两个或者两个以上下一兆 , 则写两个或 两个以上相同的下一跳。
4. 按照权利要求 1所述的实现快速重路由的方法, 其特征在于: 在所述 的步骤 c中 , 子网路由的下一跳不使用 IP地址, 关联到所述 FRR组 的 index。
5. 按照权利要求 1所述的实现快速重路由的方法, 其特征在于: 在所述 的步骤 d中 , 驱动通过调用芯片的 SDK函数, 将 FRR新的有效下一 跳 IP地址写入芯片的 ECMP表中。 按照权利要求 1所述的实现快速重路由的方法, 其特征在于: 在所述 的步骤 d中, 更新完 ECMP表的地址后, 更新所述步骤 b中的软件表。
PCT/CN2009/073649 2008-12-16 2009-08-31 实现快速重路由的方法 WO2010069182A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US13/140,054 US20110249679A1 (en) 2008-12-16 2009-08-31 Method for implementing fast reroute
ES09832864.4T ES2585401T3 (es) 2008-12-16 2009-08-31 Procedimiento para implementar un cambio de destino (reencaminado) rápido
EP09832864.4A EP2378719B1 (en) 2008-12-16 2009-08-31 Method for implementing fast reroute

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810241251.3A CN101442494B (zh) 2008-12-16 2008-12-16 一种实现快速重路由的方法
CN200810241251.3 2008-12-16

Publications (1)

Publication Number Publication Date
WO2010069182A1 true WO2010069182A1 (zh) 2010-06-24

Family

ID=40726741

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/073649 WO2010069182A1 (zh) 2008-12-16 2009-08-31 实现快速重路由的方法

Country Status (5)

Country Link
US (1) US20110249679A1 (zh)
EP (1) EP2378719B1 (zh)
CN (1) CN101442494B (zh)
ES (1) ES2585401T3 (zh)
WO (1) WO2010069182A1 (zh)

Families Citing this family (63)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101442494B (zh) * 2008-12-16 2011-06-22 中兴通讯股份有限公司 一种实现快速重路由的方法
CN102201964B (zh) * 2010-03-22 2014-02-05 杭州华三通信技术有限公司 一种实现快速路径切换的方法和装置
CN101931590A (zh) * 2010-08-27 2010-12-29 中兴通讯股份有限公司 一种路由整合方法和装置
US9191327B2 (en) 2011-02-10 2015-11-17 Varmour Networks, Inc. Distributed service processing of network gateways using virtual machines
CN102291311B (zh) * 2011-08-30 2017-03-29 中兴通讯股份有限公司 以太网接口保护方法及网络侧设备
CN102638393B (zh) * 2012-03-16 2015-08-12 中兴通讯股份有限公司 一种实现快速重路由的方法及网络节点
CN102801614B (zh) * 2012-07-17 2016-04-27 杭州华三通信技术有限公司 一种等价路由的收敛方法和网络设备
CN102984062B (zh) * 2012-11-29 2016-06-15 中兴通讯股份有限公司 一种子网路由的处理方法及报文转发设备
CN103078804B (zh) * 2012-12-28 2015-07-22 福建星网锐捷网络有限公司 等价多路径表处理方法、装置及网络设备
US9634940B2 (en) 2013-01-31 2017-04-25 Mellanox Technologies, Ltd. Adaptive routing using inter-switch notifications
CN104144120A (zh) * 2013-05-07 2014-11-12 杭州华三通信技术有限公司 转发信息配置方法及装置
US9548960B2 (en) 2013-10-06 2017-01-17 Mellanox Technologies Ltd. Simplified packet routing
US10264025B2 (en) 2016-06-24 2019-04-16 Varmour Networks, Inc. Security policy generation for virtualization, bare-metal server, and cloud computing environments
US9973472B2 (en) * 2015-04-02 2018-05-15 Varmour Networks, Inc. Methods and systems for orchestrating physical and virtual switches to enforce security boundaries
US10091238B2 (en) 2014-02-11 2018-10-02 Varmour Networks, Inc. Deception using distributed threat detection
US9729473B2 (en) 2014-06-23 2017-08-08 Mellanox Technologies, Ltd. Network high availability using temporary re-routing
US9806994B2 (en) 2014-06-24 2017-10-31 Mellanox Technologies, Ltd. Routing via multiple paths with efficient traffic distribution
US9699067B2 (en) 2014-07-22 2017-07-04 Mellanox Technologies, Ltd. Dragonfly plus: communication over bipartite node groups connected by a mesh network
CN104253746B (zh) * 2014-09-17 2017-04-12 烽火通信科技股份有限公司 一种快速重路由倒换方法及系统
US9438634B1 (en) 2015-03-13 2016-09-06 Varmour Networks, Inc. Microsegmented networks that implement vulnerability scanning
US10193929B2 (en) 2015-03-13 2019-01-29 Varmour Networks, Inc. Methods and systems for improving analytics in distributed networks
US9467476B1 (en) 2015-03-13 2016-10-11 Varmour Networks, Inc. Context aware microsegmentation
US9294442B1 (en) 2015-03-30 2016-03-22 Varmour Networks, Inc. System and method for threat-driven security policy controls
US10178070B2 (en) 2015-03-13 2019-01-08 Varmour Networks, Inc. Methods and systems for providing security to distributed microservices
US10009381B2 (en) 2015-03-30 2018-06-26 Varmour Networks, Inc. System and method for threat-driven security policy controls
US9380027B1 (en) 2015-03-30 2016-06-28 Varmour Networks, Inc. Conditional declarative policies
US9894005B2 (en) 2015-03-31 2018-02-13 Mellanox Technologies, Ltd. Adaptive routing controlled by source node
US9525697B2 (en) 2015-04-02 2016-12-20 Varmour Networks, Inc. Delivering security functions to distributed networks
US9483317B1 (en) 2015-08-17 2016-11-01 Varmour Networks, Inc. Using multiple central processing unit cores for packet forwarding in virtualized networks
US10191758B2 (en) 2015-12-09 2019-01-29 Varmour Networks, Inc. Directing data traffic between intra-server virtual machines
US9973435B2 (en) 2015-12-16 2018-05-15 Mellanox Technologies Tlv Ltd. Loopback-free adaptive routing
US9762599B2 (en) 2016-01-29 2017-09-12 Varmour Networks, Inc. Multi-node affinity-based examination for computer network security remediation
US9680852B1 (en) 2016-01-29 2017-06-13 Varmour Networks, Inc. Recursive multi-layer examination for computer network security remediation
US10819621B2 (en) 2016-02-23 2020-10-27 Mellanox Technologies Tlv Ltd. Unicast forwarding of adaptive-routing notifications
US9521115B1 (en) 2016-03-24 2016-12-13 Varmour Networks, Inc. Security policy generation using container metadata
US10178029B2 (en) 2016-05-11 2019-01-08 Mellanox Technologies Tlv Ltd. Forwarding of adaptive routing notifications
US10755334B2 (en) 2016-06-30 2020-08-25 Varmour Networks, Inc. Systems and methods for continually scoring and segmenting open opportunities using client data and product predictors
CN106603403A (zh) * 2016-08-29 2017-04-26 安徽皖通邮电股份有限公司 一种实现ip快速重路由的方法
CN106385362B (zh) * 2016-08-30 2019-08-27 迈普通信技术股份有限公司 一种非负载均衡路由切换方法及装置
US10200294B2 (en) 2016-12-22 2019-02-05 Mellanox Technologies Tlv Ltd. Adaptive routing based on flow-control credits
CN106603401B (zh) * 2017-02-16 2019-09-13 迈普通信技术股份有限公司 快速重路由方法和装置
CN109962991B (zh) * 2017-12-26 2022-06-14 中国移动通信集团四川有限公司 物联网故障处理方法、装置、设备及介质
US10644995B2 (en) 2018-02-14 2020-05-05 Mellanox Technologies Tlv Ltd. Adaptive routing in a box
US11005724B1 (en) 2019-01-06 2021-05-11 Mellanox Technologies, Ltd. Network topology having minimal number of long connections among groups of network elements
US11290493B2 (en) 2019-05-31 2022-03-29 Varmour Networks, Inc. Template-driven intent-based security
US11575563B2 (en) 2019-05-31 2023-02-07 Varmour Networks, Inc. Cloud security management
US11310284B2 (en) 2019-05-31 2022-04-19 Varmour Networks, Inc. Validation of cloud security policies
US11290494B2 (en) 2019-05-31 2022-03-29 Varmour Networks, Inc. Reliability prediction for cloud security policies
US11863580B2 (en) 2019-05-31 2024-01-02 Varmour Networks, Inc. Modeling application dependencies to identify operational risk
US11711374B2 (en) 2019-05-31 2023-07-25 Varmour Networks, Inc. Systems and methods for understanding identity and organizational access to applications within an enterprise environment
CN111130871B (zh) * 2019-12-18 2022-06-10 新华三半导体技术有限公司 保护切换方法、装置和网络设备
US11496354B2 (en) * 2020-06-16 2022-11-08 Ciena Corporation ECMP fast convergence on path failure using objects in a switching circuit
US11575594B2 (en) 2020-09-10 2023-02-07 Mellanox Technologies, Ltd. Deadlock-free rerouting for resolving local link failures using detour paths
US11411911B2 (en) 2020-10-26 2022-08-09 Mellanox Technologies, Ltd. Routing across multiple subnetworks using address mapping
US11818152B2 (en) 2020-12-23 2023-11-14 Varmour Networks, Inc. Modeling topic-based message-oriented middleware within a security system
US11876817B2 (en) 2020-12-23 2024-01-16 Varmour Networks, Inc. Modeling queue-based message-oriented middleware relationships in a security system
US12050693B2 (en) 2021-01-29 2024-07-30 Varmour Networks, Inc. System and method for attributing user behavior from multiple technical telemetry sources
US11777978B2 (en) 2021-01-29 2023-10-03 Varmour Networks, Inc. Methods and systems for accurately assessing application access risk
CN113315699B (zh) * 2021-05-17 2022-05-17 Ut斯达康通讯有限公司 一种区分优先级的多主多备快速重路由系统及方法
US11870682B2 (en) 2021-06-22 2024-01-09 Mellanox Technologies, Ltd. Deadlock-free local rerouting for handling multiple local link failures in hierarchical network topologies
US11734316B2 (en) 2021-07-08 2023-08-22 Varmour Networks, Inc. Relationship-based search in a computing environment
US11765103B2 (en) 2021-12-01 2023-09-19 Mellanox Technologies, Ltd. Large-scale network with high port utilization
CN114884868B (zh) * 2022-05-10 2024-04-12 云合智网(上海)技术有限公司 基于ecmp组的链路保护方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008016558A2 (en) * 2006-07-31 2008-02-07 Cisco Technology, Inc. Technique for multiple path forwarding of label-switched data traffic
CN101137974A (zh) * 2003-10-07 2008-03-05 思科技术公司 用于mpls快速重路由的增强切换
CN101237409A (zh) * 2008-02-27 2008-08-06 华为技术有限公司 Mpls vpn中实现快速重路由的方法及设备
CN101299722A (zh) * 2008-07-02 2008-11-05 杭州华三通信技术有限公司 一种改进的快速重路由方法和一种网络设备
CN101442494A (zh) * 2008-12-16 2009-05-27 中兴通讯股份有限公司 一种实现快速重路由的方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6778492B2 (en) * 2002-01-17 2004-08-17 Cisco Technology, Inc. Load balancing for fast reroute backup tunnels
US7230913B1 (en) * 2002-06-11 2007-06-12 Cisco Technology, Inc. MPLS fast reroute without full mesh traffic engineering
SE525800C2 (sv) * 2002-10-30 2005-04-26 Packetfront Sweden Ab Anordning och router i ett bredbandsnät som arbetar genom överföring av paketflöden, med ett av en central processorenhet omkopplat mjukvaruflöde och ett hårdvaruomkopplat accelerarat flöde
US7440415B2 (en) * 2003-05-30 2008-10-21 Ixia Virtual network addresses
US20040264503A1 (en) * 2003-06-30 2004-12-30 Microsoft Corporation Method and system for providing a virtual protocol interlayer
JP2007533172A (ja) * 2003-11-11 2007-11-15 サイトリックス ゲートウェイズ, インコーポレイテッド 偽サーバを備えた仮想プライベートネットワーク
US7680952B1 (en) * 2004-06-16 2010-03-16 Juniper Networks, Inc. Protecting connection traffic using filters
US7551551B2 (en) * 2004-12-10 2009-06-23 Cisco Technology, Inc. Fast reroute (FRR) protection at the edge of a RFC 2547 network
US7633859B2 (en) * 2005-01-26 2009-12-15 Cisco Technology, Inc. Loop prevention technique for MPLS using two labels
US7684316B2 (en) * 2008-02-12 2010-03-23 Cisco Technology, Inc. Multicast fast reroute for network topologies

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101137974A (zh) * 2003-10-07 2008-03-05 思科技术公司 用于mpls快速重路由的增强切换
WO2008016558A2 (en) * 2006-07-31 2008-02-07 Cisco Technology, Inc. Technique for multiple path forwarding of label-switched data traffic
CN101237409A (zh) * 2008-02-27 2008-08-06 华为技术有限公司 Mpls vpn中实现快速重路由的方法及设备
CN101299722A (zh) * 2008-07-02 2008-11-05 杭州华三通信技术有限公司 一种改进的快速重路由方法和一种网络设备
CN101442494A (zh) * 2008-12-16 2009-05-27 中兴通讯股份有限公司 一种实现快速重路由的方法

Also Published As

Publication number Publication date
EP2378719B1 (en) 2016-06-08
EP2378719A1 (en) 2011-10-19
ES2585401T3 (es) 2016-10-05
CN101442494B (zh) 2011-06-22
US20110249679A1 (en) 2011-10-13
CN101442494A (zh) 2009-05-27
EP2378719A4 (en) 2013-12-04

Similar Documents

Publication Publication Date Title
WO2010069182A1 (zh) 实现快速重路由的方法
JP4449903B2 (ja) ルータ装置およびネットワーク接続方式
US11032197B2 (en) Reroute detection in segment routing data plane
US7342874B2 (en) High-availability packet forwarding apparatus and method
US7126907B2 (en) Label switched communication network, a method of conditioning the network and a method of data transmission
US8055791B2 (en) Protecting connection traffic using filters
JP7176095B2 (ja) 通信方法、通信機器、および通信システム
US20060087962A1 (en) Fault tolerant network architecture
RU2574812C9 (ru) Способ и узел детектирования маршрута в сети передачи данных
EP2664105B1 (en) Method for backing up link state advertisement
CN101820395A (zh) 基于mpls的路由信息配置和私网标签添加方法及装置
US7626925B1 (en) Methods for finding a merge point node for MPLS fast re-route
JP2007067690A (ja) 最適経路選択方式
WO2013029438A1 (zh) 以太网接口保护方法及网络侧设备
JPWO2005057864A1 (ja) ネットワークの経路切替えシステム
CN101192990A (zh) 一种mpls网络中实现快速重路由的方法及设备及系统
CN100568831C (zh) 一种提高网络可靠性的方法及系统
KR100775768B1 (ko) Ⅴrrp 라우터 및 ⅴrrp 라우터간의 상태 천이 방법
US20080212610A1 (en) Communication techniques and generic layer 3 automatic switching protection
JP6490167B2 (ja) 通信装置、通信方法、コンピュータプログラムおよび通信システム
KR100431208B1 (ko) 멀티프로토콜 레이블 스위칭 시스템의 결함을 허용하는명시적 경로 관리방법 및 장치
JP4717796B2 (ja) ノード装置およびパス設定方法
JP2018019232A (ja) パケット伝送装置、及び、経路切替制御方法
CN112689032A (zh) 一种数据处理方法、设备及存储介质
JP2008060995A (ja) Mplsネットワーク転送制御方法及びシステム

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: 09832864

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 13140054

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 5246/DELNP/2011

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 2009832864

Country of ref document: EP