公开(公告)号：US06956849B1

公开(公告)日：2005-10-18

申请号：US09728908

申请日：2000-12-01

Applicant: Man Pak Yip ,  Vijayaraghavan Doraiswami

Inventor： Man Pak Yip ,  Vijayaraghavan Doraiswami

IPC: H04L12/50 ,  H04L12/56

CPC classification number: H04L47/822 ,  H04L47/39 ,  H04L47/70 ,  H04L47/765 ,  H04L47/801 ,  H04L47/805

Abstract: A line card within a switching node coupled to a network is described. The line card includes a link interface for transmitting communications along a communication link within the network. The link interface further includes multiple logical interfaces having one or more partitions and one or more buffers. The partitions and the buffers accommodate multiple classes of service requirements for the communications transmitted within the network. In one embodiment, bandwidth management and connection admission control in a network encompasses both the line card and the system which uses the line card.

Abstract translation: 描述了耦合到网络的交换节点内的线路卡。 线卡包括用于沿着网络内的通信链路发送通信的链路接口。 链路接口还包括具有一个或多个分区和一个或多个缓冲器的多个逻辑接口。 分区和缓冲器适用于在网络内传输的通信的多种服务要求。 在一个实施例中，网络中的带宽管理和连接准入控制包括线路卡和使用线路卡的系统。

公开(公告)号：US20080276004A1

公开(公告)日：2008-11-06

申请号：US11742971

申请日：2007-05-01

Applicant: Allan Thomson ,  David S. Stephenson ,  Subhasri Dhesikan ,  Vijayaraghavan Doraiswami ,  Shree Murthy

Inventor： Allan Thomson ,  David S. Stephenson ,  Subhasri Dhesikan ,  Vijayaraghavan Doraiswami ,  Shree Murthy

IPC: H04L12/28 ,  G06F17/00

CPC classification number: H04W4/02 ,  H04L29/12226 ,  H04L61/2015 ,  H04L67/18

Abstract: A method for populating location wiremap databases. In particular implementations, a method includes establishing a link layer connection with a client on a switch port, where the switch port is associated with a port identifier and is mapped to a location; identifying one or more connection attributes of the connection, where the connection attributes comprise a network layer address of the client; and transmitting the port identifier and the network layer address of the client to a location server.

Abstract translation: 填充位置线图数据库的方法。 在特定实现中，一种方法包括与交换机端口上的客户端建立链路层连接，其中交换机端口与端口标识符相关联并被映射到一个位置; 识别所述连接的一个或多个连接属性，其中所述连接属性包括所述客户端的网络层地址; 并将客户端的端口标识符和网络层地址发送到位置服务器。

公开(公告)号：US08838831B2

公开(公告)日：2014-09-16

申请号：US11742971

申请日：2007-05-01

Applicant: Allan Thomson ,  David S. Stephenson ,  Subhasri Dhesikan ,  Vijayaraghavan Doraiswami ,  Shree Murthy

Inventor： Allan Thomson ,  David S. Stephenson ,  Subhasri Dhesikan ,  Vijayaraghavan Doraiswami ,  Shree Murthy

IPC: H04L12/58 ,  H04L29/08 ,  H04W4/02 ,  H04L29/12

CPC classification number: H04W4/02 ,  H04L29/12226 ,  H04L61/2015 ,  H04L67/18

Abstract: A method for populating location wiremap databases. In particular implementations, a method includes establishing a link layer connection with a client on a switch port, where the switch port is associated with a port identifier and is mapped to a location; identifying one or more connection attributes of the connection, where the connection attributes comprise a network layer address of the client; and transmitting the port identifier and the network layer address of the client to a location server.

Abstract translation: 填充位置线图数据库的方法。 在特定实现中，一种方法包括与交换机端口上的客户端建立链路层连接，其中交换机端口与端口标识符相关联并被映射到一个位置; 识别所述连接的一个或多个连接属性，其中所述连接属性包括所述客户端的网络层地址; 并将客户端的端口标识符和网络层地址发送到位置服务器。

公开(公告)号：US08189588B2

公开(公告)日：2012-05-29

申请号：US12403906

申请日：2009-03-13

Applicant: Navindra Yadav ,  Shree Murthy ,  Vijayaraghavan Doraiswami ,  Peter Geoffrey Jones

Inventor： Navindra Yadav ,  Shree Murthy ,  Vijayaraghavan Doraiswami ,  Peter Geoffrey Jones

IPC: H04L12/56

CPC classification number: H04L67/1095 ,  H04L45/00 ,  H04L45/24 ,  H04L45/28 ,  H04L69/28

Abstract: In an example embodiment, packets for a selected flow are replicated and sent over one or more diverse paths, such as a primary path and at least one secondary path, to a destination switching device. At the destination switching device, one copy of the replicated packets is selected for delivery to the destination, and the remaining copies are discarded. In the event that packets are not received at the destination switching device due to loss of connection on the primary path or packets are not timely delivered due to congestion on the primary path, a different path may be selected as the primary path.

Abstract translation: 在示例实施例中，用于所选流的分组被复制并通过一个或多个不同路径（诸如主路径和至少一个次要路径）发送到目的地交换设备。 在目的地交换设备处，选择复制的分组的一个副本以传送到目的地，并且剩余的副本被丢弃。 如果由于主路径上的连接丢失而在目的地交换设备处没有接收到分组，否则由于主路径上的拥塞而不能及时地传送分组，可以选择不同的路径作为主要路径。

公开(公告)号：US06941388B1

公开(公告)日：2005-09-06

申请号：US09746706

申请日：2000-12-22

Applicant: Vijayaraghavan Doraiswami

Inventor： Vijayaraghavan Doraiswami

IPC: G06F3/00 ,  H04L12/66

CPC classification number: H04L12/66

Abstract: A standby card is powered after an active card. A connection command is received by the standby card from the active card. The connection command is associated with a logical connection number (LCN) for a connection. The LCN is used as a first index to a location in a first memory area to retrieve a second index to a location in a second memory area. The second index is used to access the connection from the location in the second memory area.

Abstract translation: 备用卡在活动卡上供电。 备用卡从活动卡接收到连接命令。 连接命令与连接的逻辑连接号（LCN）相关联。 LCN用作第一存储器区域中的位置的第一索引，以便检索到第二存储区域中的位置的第二索引。 第二个索引用于从第二个存储区域中的位置访问连接。

公开(公告)号：US09015438B2

公开(公告)日：2015-04-21

申请号：US13692622

申请日：2012-12-03

Applicant: Ramsundar Janakiraman ,  Prasad Palkar ,  Brijesh Nambiar ,  Sridhar Kamsetty ,  Vijayaraghavan Doraiswami

Inventor： Ramsundar Janakiraman ,  Prasad Palkar ,  Brijesh Nambiar ,  Sridhar Kamsetty ,  Vijayaraghavan Doraiswami

IPC: G06F12/00 ,  G06F9/52 ,  H04L12/54

CPC classification number: G06F9/526 ,  H04L12/56

Abstract: The present disclosure discloses a method and network device for achieving enhanced performance with multiple CPU cores in a network device having a symmetric multiprocessing architecture. The disclosed method allows for storing, by each central processing unit (CPU) core, a non-atomic data structure, which is specific to each networking CPU core, in a memory shared by the plurality of CPU cores. Also, the memory is not associated with any locking mechanism. In response to a data packet is received by a particular CPU core, the disclosed system will update a value of the non-atomic data structure corresponding to the particular CPU core. The data structure may be a counter or a fragment table. Further, a dedicated CPU core is allocated to process only data packets received from other CPU cores, and is responsible for dynamically responding to queries receives from a control plane process.

Abstract translation: 本公开公开了一种用于在具有对称多处理架构的网络设备中实现具有多个CPU内核的增强性能的方法和网络设备。 所公开的方法允许由每个中央处理单元（CPU）存储由多个CPU核共享的存储器中的每个联网CPU核心特有的非原子数据结构。 此外，内存不与任何锁定机制相关联。 响应于特定CPU核心接收到的数据分组，所公开的系统将更新对应于特定CPU核心的非原子数据结构的值。 数据结构可以是计数器或片段表。 此外，分配专用CPU核以仅处理从其他CPU核心接收的数据分组，并且负责动态响应从控制平面进程接收到的查询。

公开(公告)号：US20140156954A1

公开(公告)日：2014-06-05

申请号：US13692622

申请日：2012-12-03

Applicant: Ramsundar Janakiraman ,  Prasad Palkar ,  Brijesh Nambiar ,  Sridhar Kamsetty ,  Vijayaraghavan Doraiswami

Inventor： Ramsundar Janakiraman ,  Prasad Palkar ,  Brijesh Nambiar ,  Sridhar Kamsetty ,  Vijayaraghavan Doraiswami

IPC: G06F12/14

CPC classification number: G06F9/526 ,  H04L12/56

Abstract: The present disclosure discloses a method and network device for achieving enhanced performance with multiple CPU cores in a network device having a symmetric multiprocessing architecture. The disclosed method allows for storing, by each central processing unit (CPU) core, a non-atomic data structure, which is specific to each networking CPU core, in a memory shared by the plurality of CPU cores. Also, the memory is not associated with any locking mechanism. In response to a data packet is received by a particular CPU core, the disclosed system will update a value of the non-atomic data structure corresponding to the particular CPU core. The data structure may be a counter or a fragment table. Further, a dedicated CPU core is allocated to process only data packets received from other CPU cores, and is responsible for dynamically responding to queries receives from a control plane process.

Abstract translation: 本公开公开了一种用于在具有对称多处理架构的网络设备中实现具有多个CPU内核的增强性能的方法和网络设备。 所公开的方法允许由每个中央处理单元（CPU）存储由多个CPU核共享的存储器中的每个联网CPU核心特有的非原子数据结构。 此外，内存不与任何锁定机制相关联。 响应于特定CPU核心接收到的数据分组，所公开的系统将更新对应于特定CPU核心的非原子数据结构的值。 数据结构可以是计数器或片段表。 此外，分配专用CPU核以仅处理从其他CPU核心接收的数据分组，并且负责动态响应从控制平面进程接收到的查询。

公开(公告)号：US20100235593A1

公开(公告)日：2010-09-16

申请号：US12403906

申请日：2009-03-13

Applicant: Navindra YADAV ,  Shree Murthy ,  Vijayaraghavan Doraiswami ,  Peter Geoffrey Jones

Inventor： Navindra YADAV ,  Shree Murthy ,  Vijayaraghavan Doraiswami ,  Peter Geoffrey Jones

IPC: G06F12/16

CPC classification number: H04L67/1095 ,  H04L45/00 ,  H04L45/24 ,  H04L45/28 ,  H04L69/28

Abstract: In an example embodiment, packets for a selected flow are replicated and sent over one or more diverse paths, such as a primary path and at least one secondary path, to a destination switching device. At the destination switching device, one copy of the replicated packets is selected for delivery to the destination, and the remaining copies are discarded. In the event that packets are not received at the destination switching device due to loss of connection on the primary path or packets are not timely delivered due to congestion on the primary path, a different path may be selected as the primary path.

Abstract translation: 在示例实施例中，用于所选流的分组被复制并通过一个或多个不同路径（诸如主路径和至少一个次要路径）发送到目的地交换设备。 在目的地交换设备处，选择复制的分组的一个副本以传送到目的地，并且剩余的副本被丢弃。 如果由于主路径上的连接丢失而在目的地交换设备处没有接收到分组，否则由于主路径上的拥塞而不能及时地传送分组，可以选择不同的路径作为主要路径。