公开(公告)号：US20090147674A1

公开(公告)日：2009-06-11

申请号：US12348766

申请日：2009-01-05

Applicant: John Galen Scudder ,  David D. Ward ,  Clarence Filsfils

Inventor： John Galen Scudder ,  David D. Ward ,  Clarence Filsfils

IPC: G06F11/00

CPC classification number: H04L45/22 ,  H04L45/02 ,  H04L45/28 ,  H04L45/50 ,  H04L47/2408

Abstract: In one embodiment, an edge device communicates with a neighboring routing domain. A failure that prevents communication between the edge device and the neighboring routing is detected. When the edge device thereafter receives a data packet that is directed to the neighboring routing domain, it determines if the received data packet was rerouted to the edge device from another edge device coupled to the neighboring routing domain. If the received data packet was not rerouted to the edge device from another edge device coupled to the neighboring routing domain, the edge device reroutes the received data packet to another edge device for forwarding to the neighboring routing domain. However, if the received data packet was rerouted to the edge device from another edge device coupled to the neighboring routing domain, the edge device prevents the received data packet from being rerouted a second time to prevent loops.

Abstract translation: 在一个实施例中，边缘设备与相邻路由域通信。 检测到阻止边缘设备和相邻路由之间的通信的故障。 当边缘设备此后接收到指向相邻路由域的数据分组时，它确定所接收的数据分组是否从耦合到相邻路由域的另一边缘设备重新路由到边缘设备。 如果接收到的数据分组没有从耦合到相邻路由域的另一边缘设备重新路由到边缘设备，则边缘设备将接收到的数据分组重新路由到另一个边缘设备，以转发到相邻路由域。 然而，如果接收到的数据分组从耦合到相邻路由域的另一边缘设备重新路由到边缘设备，则边缘设备第二次防止接收到的数据分组被重新路由以防止环路。

公开(公告)号：US07508772B1

公开(公告)日：2009-03-24

申请号：US11445845

申请日：2006-06-02

Applicant: David D. Ward ,  Chandrashekhar Appanna

Inventor： David D. Ward ,  Chandrashekhar Appanna

IPC: H04L12/28 ,  H04L12/56

CPC classification number: H04L45/22 ,  H04L45/02 ,  H04L45/04 ,  H04L45/28 ,  H04L45/54

Abstract: A method and apparatus for partial graceful restart in Border Gateway Protocol (BGP) are disclosed. A first BGP host detects that a first BGP session established between the first BGP host and a second BGP host has become unavailable. In a Forwarding Information Base (FIB) at the first BGP host, a first set of one or more routes to one or more address destinations is replaced with a second set of one or more routes to the one or more address destinations. The first set of one or more routes indicates that the one or more address destinations are reachable through the second BGP host. The second set of one or more routes indicates that the one or more address destinations are reachable through a third BGP host. The first BGP host receives a BGP OPEN message from the second BGP host to open a new BGP session between the first BGP host and the second BGP host. In response to receiving the BGP OPEN message, the first set of one or more routes is re-installed in the FIB at the first BGP host.

Abstract translation: 公开了一种在边界网关协议（BGP）中进行局部优雅重启的方法和装置。 第一个BGP主机检测到第一个BGP主机和第二个BGP主机之间建立的第一个BGP会话变得不可用。 在第一BGP主机的转发信息库（FIB）中，向一个或多个地址目的地的第一组一条或多条路由替换到一个或多个地址目的地的一条或多条路由。 第一组一个或多个路由表示一个或多个地址目的地可通过第二个BGP主机到达。 第一组一个或多个路由表示一个或多个地址目的地可通过第三个BGP主机到达。 第一个BGP主机从第二个BGP主机接收到一个BGP OPEN消息，以在第一个BGP主机和第二个BGP主机之间打开一个新的BGP会话。 响应于接收到BGP OPEN消息，第一组BGP路由重新安装在第一台BGP主机的FIB中。

公开(公告)号：US07471636B2

公开(公告)日：2008-12-30

申请号：US11063803

申请日：2005-02-23

Applicant: Jean-Philippe Vasseur ,  Stefano Previdi ,  James N. Guichard ,  David D. Ward

Inventor： Jean-Philippe Vasseur ,  Stefano Previdi ,  James N. Guichard ,  David D. Ward

IPC: H04L12/26

CPC classification number: H04L12/56

Abstract: A data communication device (e.g., a router) originates a network configuration message in response to a network topology change or so as to refresh a configuration message. The data communication device encodes a timestamp in the network configuration message. The timestamp indicates a time of originating the network configuration message. Further, the data communication device transmits the network configuration message over the network to other network devices that, in turn, initiate further broadcast of at least a portion of contents of the network configuration message. Based on the timestamp of the network configuration message, the data communication devices receiving the network configuration message identify transmission time value indicating how long the network configuration message takes to be conveyed over the network to the other network devices. The data communication devices utilize the transmission time value as a timeout period for determining whether a data communication device failure occurs.

Abstract translation: 数据通信设备（例如，路由器）响应于网络拓扑变化而发起网络配置消息，以便刷新配置消息。 数据通信设备对网络配置消息中的时间戳进行编码。 时间戳表示发起网络配置消息的时间。 此外，数据通信设备通过网络将网络配置消息发送到其他网络设备，其进一步发起网络配置消息的至少一部分内容的进一步广播。 基于网络配置消息的时间戳，接收网络配置消息的数据通信设备识别表示网络配置消息通过网络传送到其他网络设备多长时间的传输时间值。 数据通信设备利用传输时间值作为用于确定数据通信设备故障是否发生的超时周期。

公开(公告)号：US20080253381A1

公开(公告)日：2008-10-16

申请号：US11786506

申请日：2007-04-12

Applicant: David D. Ward ,  Luca Martini ,  Jim Guichard

Inventor： David D. Ward ,  Luca Martini ,  Jim Guichard

IPC: H04L12/28

CPC classification number: H04L45/24 ,  H04L41/0668 ,  H04L45/10 ,  H04L45/28 ,  H04L45/68 ,  H04L47/10 ,  H04L47/125 ,  H04L47/26

Abstract: In one embodiment, a communications distribution process maintains at least two pseudowires through a network such that the pseudowires share a burden of delivering data through the network. The communications distribution process receives feedback data concerning operation of each pseudowire. The communications distribution process utilizes the feedback data to distribute communications to the common destination across each of the pseudowires. Additionally, the communications distribution process utilizes the feedback to establish at least one new pseudowire, in addition to the first pseudowire and the second pseudowire, for transmission of data traffic.

Abstract translation: 在一个实施例中，通信分发过程通过网络维持至少两个伪线，使得伪线共享通过网络传送数据的负担。 通信分配过程接收关于每个伪线的操作的反馈数据。 通信分配过程利用反馈数据将通信分配到每个伪线上的公共目的地。 此外，通信分配过程利用反馈来建立除了第一伪线和第二伪线之外的至少一个新的伪线，用于数据业务的传输。

公开(公告)号：US20070208874A1

公开(公告)日：2007-09-06

申请号：US11449162

申请日：2006-06-08

Applicant: Stefano B. Previdi ,  David D. Ward

Inventor： Stefano B. Previdi ,  David D. Ward

IPC: G06F15/173

CPC classification number: H04L45/70 ,  H04L67/1002 ,  H04L67/1008 ,  H04L67/101 ,  H04L67/2814 ,  H04L67/32

Abstract: A technique optimizes routing of application data streams on an Internet Protocol (IP) backbone in a computer network. According to the novel technique, a client router learns of server states (e.g., number of pending requests, etc.) of a plurality of application servers and also determines metrics of intermediate links between the application servers and the client router (intermediate link metrics), e.g., particularly link metrics in a direction from the application servers to the client router. Upon receiving an application request from an application client (“client request”), the client router determines to which of the application servers the client request is to be sent based on the server states and intermediate link metrics, and sends the client request accordingly.

Abstract translation: 一种技术优化了计算机网络中的因特网协议（IP）骨干网上的应用数据流的路由。 根据新技术，客户端路由器学习多个应用服务器的服务器状态（例如，等待请求的数量等），并且还确定应用服务器与客户端路由器之间的中间链路的度量（中间链路度量） ，例如，特别是从应用服务器到客户端路由器的方向上的链路度量。 在从应用客户端（“客户端请求”）接收到应用请求时，客户端路由器根据服务器状态和中间链路度量确定哪个应用服务器要发送客户端请求，并相应地发送客户端请求。

公开(公告)号：US07023808B2

公开(公告)日：2006-04-04

申请号：US10743973

申请日：2003-12-23

Applicant: David Alexander Ball ,  R. Eric Bennett ,  Martin Hesketh ,  John Galen Scudder ,  David D. Ward

Inventor： David Alexander Ball ,  R. Eric Bennett ,  Martin Hesketh ,  John Galen Scudder ,  David D. Ward

IPC: H04J1/16

CPC classification number: H04L45/56 ,  H04L45/04 ,  H04L45/12 ,  H04L45/44

Abstract: A partial best path technique distributes route selection in a routing protocol implementation on a router. The technique also ensures that announced paths received from peers of the router (i.e., a “load”) are compared in a correct order to select best paths that are then used by the router to forward packets and to advertise to the peers. When employed in a distributed architecture, the technique further reduces memory usage. To that end, the partial best path technique enhances a best path selection algorithm executed by the router to enable dispersion of the received path load among processing nodes or elements of the router, while maintaining the ordering requirement of the algorithm. The partial best path technique essentially provides an enhancement to the best path selection algorithm that selects a subset of paths from a plurality of paths, with that subset being the minimal subset needed to select the best paths.

Abstract translation: 部分最佳路径技术在路由器上的路由协议实现中分配路由选择。 该技术还确保从路由器的对等端（即，“负载”）接收到的公告路径以正确的顺序进行比较，以选择最佳路径，然后由路由器用于转发数据包并向对等体通告。 当在分布式架构中使用时，该技术进一步减少内存使用。 为此，部分最佳路径技术增强了由路由器执行的最佳路径选择算法，以便在维护该算法的排序要求的同时，使路由器的处理节点或元素之间的接收路径负载分散。 部分最佳路径技术基本上为最佳路径选择算法提供了增强，该算法从多个路径中选择路径的子集，该子集是选择最佳路径所需的最小子集。

公开(公告)号：US08825898B2

公开(公告)日：2014-09-02

申请号：US11449162

申请日：2006-06-08

Applicant: Stefano B. Previdi ,  David D. Ward

Inventor： Stefano B. Previdi ,  David D. Ward

IPC: G06F15/173 ,  H04L29/08 ,  G06F15/16

CPC classification number: H04L45/70 ,  H04L67/1002 ,  H04L67/1008 ,  H04L67/101 ,  H04L67/2814 ,  H04L67/32

Abstract: A technique optimizes routing of application data streams on an Internet Protocol (IP) backbone in a computer network. According to the novel technique, a client router learns of server states (e.g., number of pending requests, etc.) of a plurality of application servers and also determines metrics of intermediate links between the application servers and the client router (intermediate link metrics), e.g., particularly link metrics in a direction from the application servers to the client router. Upon receiving an application request from an application client (“client request”), the client router determines to which of the application servers the client request is to be sent based on the server states and intermediate link metrics, and sends the client request accordingly.

Abstract translation: 一种技术优化了计算机网络中的因特网协议（IP）骨干网上的应用数据流的路由。 根据新技术，客户端路由器学习多个应用服务器的服务器状态（例如，等待请求的数量等），并且还确定应用服务器与客户端路由器之间的中间链路的度量（中间链路度量） ，例如，特别是从应用服务器到客户端路由器的方向上的链路度量。 在从应用客户端（“客户端请求”）接收到应用请求时，客户端路由器根据服务器状态和中间链路度量确定哪个应用服务器要发送客户端请求，并相应地发送客户端请求。

公开(公告)号：US08264962B2

公开(公告)日：2012-09-11

申请号：US11168694

申请日：2005-06-27

Applicant: Jean-Philippe Vasseur ,  David D. Ward ,  Stefano B. Previdi ,  Clarence Filsfils

Inventor： Jean-Philippe Vasseur ,  David D. Ward ,  Stefano B. Previdi ,  Clarence Filsfils

IPC: G01R31/08

CPC classification number: H04L45/22 ,  H04L45/02 ,  H04L45/28 ,  H04L47/70 ,  H04L47/746 ,  H04L47/762 ,  H04L47/825 ,  H04L47/829

Abstract: A technique dynamically resizes Traffic Engineering (TE) Label Switched Paths (LSPs) at a head-end node of the TE-LSPs in preparation to receive redirected traffic in response to an event in a computer network. The novel dynamic TE-LSP resizing technique is based on the detection of an event in the network that could cause traffic destined for one or more other (“remote”) head-end nodes of one or more TE-LSPs to be redirected to an event-detecting (“local”) head-end node of one or more TE-LSPs. An example of such a traffic redirection event is failure of a remote head-end node or failure of any of its TE-LSPs. Specifically, the local head-end node maintains TE-LSP steady state sampling and resizing frequencies to adapt the bandwidth of its TE-LSP(s) to gradual changes in the network over time. Upon detection of an event identifying possible traffic redirection, the local head-end node enters a Fast Resize (FR) state, in which the sampling and resizing frequencies are increased to quickly adapt the TE-LSP bandwidth(s) to any received redirected traffic.

Abstract translation: 技术动态地调整TE-LSP的头端节点处的流量工程（TE）标签交换路径（LSP），准备响应于计算机网络中的事件接收重定向的流量。 新型动态TE-LSP调整大小技术是基于网络中可能导致一个或多个TE-LSP的一个或多个其他（“远程”）头端节点的流量被重定向到的事件的检测。 一个或多个TE-LSP的事件检测（“本地”）头端节点。 这种流量重定向事件的示例是远程头端节点的故障或其任何TE-LSP的故障。 具体来说，本地前端节点维护TE-LSP稳态采样和调整频率，以适应其TE-LSP的带宽随时间逐渐变化的网络。 在检测到识别可能的业务重定向的事件时，本地前端节点进入快速调整大小（FR）状态，其中增加采样和调整大小频率以快速地将TE-LSP带宽适配到任何接收到的重定向业务 。

公开(公告)号：US08082340B2

公开(公告)日：2011-12-20

申请号：US11343051

申请日：2006-01-30

Applicant: Clarence Filsfils ,  David D. Ward

Inventor： Clarence Filsfils ,  David D. Ward

IPC: G06F15/173

CPC classification number: H04L45/00 ,  H04L12/40 ,  H04L12/413 ,  H04L41/069 ,  H04L41/12 ,  H04L43/00 ,  H04L45/28

Abstract: A technique distinguishes between link and node failure using bidirectional forwarding detection (BFD) in a computer network. According to the novel technique, a BFD session is established over a first link between a monitoring node and a first interface of a monitored node. In addition, one or more novel unidirectional BFD (uBFD) sessions from one or more corresponding other interfaces of the monitored node to the monitoring node (e.g., as configured by the monitored node to detect its failure), the one or more uBFD sessions traversing links other than the first link (e.g., “not-via” the monitored node) are established, the one or more uBFD sessions traversing links other than the first link (e.g., “not-via” the monitored node). For instance, the one or more uBFD sessions correspond to one or more line cards of the monitored node. In response to detecting a lack of connectivity resulting from, e.g., failure of the BFD session, the monitoring node determines that the first link has failed if at least one uBFD session is operational. Moreover, in response to detecting failure of the BFD session and all of the one or more uBFD sessions, the monitoring node determines that the monitored node has failed.

Abstract translation: 一种技术在计算机网络中使用双向转发检测（BFD）来区分链路和节点故障。 根据这种新技术，在被监控节点的监控节点和第一个接口之间的第一个链路上建立BFD会话。 另外，一个或多个新颖的单向BFD（uBFD）会话从受监视节点的一个或多个对应的其他接口到监视节点（例如，由监视节点配置以检测其故障），一个或多个uBFD会话遍历 建立除第一链路之外的其他链路（例如，“未通过”的被监视节点），一个或多个遍历第一链路以外的链路的uBFD会话（例如，“不通过被监控节点”）。 例如，一个或多个uBFD会话对应于被监视节点的一个或多个线路卡。 响应于检测到由于例如BFD会话的故障而导致的连接的缺乏，如果至少一个uBFD会话可操作，则监视节点确定第一链路已经失败。 此外，响应于检测到BFD会话和所有一个或多个uBFD会话的故障，监视节点确定被监控节点已经失败。

公开(公告)号：US20110213890A1

公开(公告)日：2011-09-01

申请号：US13098627

申请日：2011-05-02

Applicant: David D. Ward ,  John Scudder

Inventor： David D. Ward ,  John Scudder

IPC: G06F15/16

CPC classification number: H04L45/28 ,  H04L45/04 ,  H04L45/58 ,  H04L45/60 ,  H04L69/16 ,  H04L69/163 ,  H04L69/40

Abstract: A method and apparatus for recovering from a controlled failover of a BGP speaker is provided. A user sends, to a network element, a request to switch a designation of an active Border Gateway Protocol (BGP) speaker of the network element from a first BGP speaker to a second BGP speaker. After receiving the request, the network element pauses operation of a transport for BGP. Routing data, which describes a state of a first routing information base (RIB) maintained by the active BGP speaker, is transferred from the first BGP speaker to the second BGP speaker. Thereafter, the first BGP speaker may instruct the second BGP speaker to become the active BGP speaker. After the second BGP speaker becomes the active BGP speaker, the second BGP speaker resumes operation of the BGP transport.

Abstract translation: 提供了一种用于从BGP扬声器的受控故障转移中恢复的方法和装置。 用户向网络单元发送将网元的主动边界网关协议（BGP）扬声器的指定从第一BGP扬声器切换到第二BGP扬声器的请求。 接收到请求后，网元暂停运行BGP。 描述由主动BGP扬声器维护的第一路由信息库（RIB）的状态的路由数据从第一BGP发言者传送到第二BGP发言者。 此后，第一个BGP扬声器可以指示第二个BGP扬声器成为主动的BGP扬声器。 第二个BGP扬声器成为主动BGP扬声器后，第二个BGP扬声器恢复BGP传输的运行。