利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

172 条记录 (耗时 0.027 秒)

    System and method for bypassing an output queue structure of a switch
    1.
    发明授权
    System and method for bypassing an output queue structure of a switch 有权
    绕过交换机的输出队列结构的系统和方法

    公开(公告)号：US07826468B2

    公开(公告)日：2010-11-02

    申请号：US11462402

    申请日：2006-08-04

    申请人： Takeshi Shimizu ,  Yukihiro Nakagawa

    发明人： Takeshi Shimizu ,  Yukihiro Nakagawa

    IPC分类号： H04L12/28

    CPC分类号： H04L47/6205 ,  H04L47/50 ,  H04L47/527 ,  H04L49/254 ,  H04L49/3036

    摘要： In particular embodiments of the present invention, a method for bypassing an output queue structure of a switch is provided. In a particular embodiment, a method for bypassing an output queue structure of a switch includes receiving a packet at an input port of a switch, storing the packet in a memory of the switch, and generating a forwarding request associated with the stored packet and with a particular output port of the switch, the forwarding request identifying the location in the memory of the stored packet. The method also includes determining whether to cause the forwarding request to bypass an output queue structure associated with the particular output port. The method further includes, using the forwarding request, retrieving from the memory the packet associated with the forwarding request and sending the retrieved packet from the particular output port.

    摘要翻译： 在本发明的特定实施例中，提供了一种绕过交换机的输出队列结构的方法。 在特定实施例中，用于绕过交换机的输出队列结构的方法包括在交换机的输入端口处接收分组，将分组存储在交换机的存储器中，并且生成与所存储的分组相关联的转发请求以及与 交换机的特定输出端口，转发请求标识存储的分组的存储器中的位置。 该方法还包括确定是否使转发请求绕过与特定输出端口相关联的输出队列结构。 所述方法还包括：使用所述转发请求，从所述存储器检索与所述转发请求相关联的分组，并从所述特定输出端口发送所检索的分组。

    Architecture for switching packets in a high-speed switching environment
    3.
    发明授权
    Architecture for switching packets in a high-speed switching environment 有权
    用于在高速交换环境中切换数据包的架构

    公开(公告)号：US07242684B2

    公开(公告)日：2007-07-10

    申请号：US10360094

    申请日：2003-02-07

    申请人： Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： H04B3/20

    CPC分类号： H04L49/103 ,  H04L49/201 ,  H04L49/3009 ,  H04L49/3036 ,  H04L49/35

    摘要： A system for switching packets in a high-speed switching environment includes one or more memory structures, multiple input structures that can each write to each of the one or more memory structures, and a first switching structure that couples the input structures to the one or more memory structures. The system also includes multiple output structures that can each read from each of the one or more memory structures and communicate a first portion of a packet to a first component of a communications network before an input structure has received a second portion of the packet from a second component of the communications network. The system also includes a second switching structure that couples the plurality of output structures to the one or more memory structures. The second switching structure is coupled to the one or more memory structures by a first number of links and coupled to the plurality of output structures by a second number of links, and the first number of links is twice or more the second number of links.

    摘要翻译： 用于在高速切换环境中切换分组的系统包括一个或多个存储器结构，每个可以写入一个或多个存储器结构中的每一个的多个输入结构，以及将输入结构耦合到一个或多个存储器结构的第一开关结构， 更多的内存结构。 该系统还包括多个输出结构，每个输出结构可以从一个或多个存储器结构中的每一个读取，并且在输入结构已经从一个或多个存储器结构接收到分组的第二部分之前将分组的第一部分传送到通信网络的第一组件 通信网络的第二个组件。 该系统还包括将多个输出结构耦合到一个或多个存储器结构的第二切换结构。 第二开关结构通过第一数量的链路耦合到一个或多个存储器结构，并且通过第二数量的链路耦合到多个输出结构，并且第一数量的链路是第二数量的链路的两倍或更多。

    Managing shared memory resources in a high-speed switching environment
    4.
    发明授权
    Managing shared memory resources in a high-speed switching environment 有权
    在高速交换环境中管理共享内存资源

    公开(公告)号：US07650413B2

    公开(公告)日：2010-01-19

    申请号：US10360085

    申请日：2003-02-07

    申请人： Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： G06F15/173

    CPC分类号： G06F9/5016 ,  Y02D10/22

    摘要： A system for managing shared memory resources in a high-speed switching environment includes a data memory logically divided into multiple blocks that correspond to multiple credits. The system includes a central agent that can maintain a pool of available credits that includes one or more of the credits. The central agent can also allocate a credit to an input port module from the pool of credits. The allocated credit indicates that the corresponding one of the blocks is available to the input port module. The system includes multiple input port modules that can each track allocated credits received from the central agent, receive a packet, determine whether enough of the allocated credits are available to write the packet to the data memory, and, if enough of the allocated credits are available, write the packet to one or more blocks corresponding to one or more of the allocated credits that are available. The system includes multiple output port modules that can each read from a block and return the credit corresponding to the read block to the pool of credits.

    摘要翻译： 用于在高速交换环境中管理共享存储器资源的系统包括逻辑上划分为对应于多个信用的多个块的数据存储器。 该系统包括可以维护包括一个或多个信用的可用信用量池的中央代理。 中央代理人还可以从信用卡分配信用到输入端口模块。 分配的信用表示对应的一个块对输入端口模块可用。 该系统包括多个输入端口模块，每个输入端口模块可以各自跟踪从中央代理接收到的分配的信用，接收分组，确定足够的分配的信用是否可用于将数据包写入数据存储器，并且如果分配的信用量足够 将分组写入与可用的一个或多个所分配的信用相对应的一个或多个块。 该系统包括多个输出端口模块，每个输出端口模块可以从块读取并将与读取块相对应的信用返回到信用卡。

    System and Method for Filtering Packets in a Switching Environment
    5.
    发明申请
    System and Method for Filtering Packets in a Switching Environment 有权
    在切换环境中过滤数据包的系统和方法

    公开(公告)号：US20080123525A1

    公开(公告)日：2008-05-29

    申请号：US11462513

    申请日：2006-08-04

    申请人： Takashi Miyoshi ,  Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Takashi Miyoshi ,  Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： H04L12/56

    CPC分类号： H04L49/555 ,  H04L49/251 ,  H04L49/254 ,  H04L49/35 ,  H04L49/358 ,  H04L49/602

    摘要： In particular embodiments of the present invention, a method for filtering packets in a switching environment is provided. In particular embodiments, the method includes receiving a packet at an input port of a switch, the switch comprising a memory and one or more output ports. The method also includes storing at least a portion of the packet in the memory and determining one or more output ports from which the packet is to be communicated from the switch. The method further includes, after beginning to determine one or more output ports from which the packet is to be communicated from the switch, determining whether the packet is an illegal packet. The method also includes, if the packet is an illegal packet, dropping the packet from the memory, and if the packet is a legal packet, communicating the packet from the determined one or more output ports.

    摘要翻译： 在本发明的特定实施例中，提供了一种用于在交换环境中过滤分组的方法。 在特定实施例中，该方法包括在交换机的输入端口处接收分组，该交换机包括存储器和一个或多个输出端口。 所述方法还包括将所述分组的至少一部分存储在所述存储器中并确定要从所述交换机传送分组的一个或多个输出端口。 该方法还包括：在开始确定要从该交换机传送分组的一个或多个输出端口之后，确定该分组是否是非法分组。 该方法还包括：如果分组是非法分组，则丢弃来自存储器的分组，并且如果分组是合法分组，则从确定的一个或多个输出端口传送分组。

    Multicasting in a high-speed switching environment
    8.
    发明授权
    Multicasting in a high-speed switching environment 有权
    在高速交换环境中组播

    公开(公告)号：US07447201B2

    公开(公告)日：2008-11-04

    申请号：US10360078

    申请日：2003-02-07

    申请人： Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： H04L12/28 ,  H04L12/56

    CPC分类号： H04L49/201 ,  H04L49/251 ,  H04L49/90

    摘要： A system for multicasting in a high-speed switching environment includes a data memory logically divided into a plurality of blocks, a plurality of output ports, and a central agent. The central agent can determine whether all one or more output ports for a packet stored in one or more blocks have read the packet. Each block of the packet can be added to a pool of available blocks when a last one of the one or more output ports for the packet has read from the block.

    摘要翻译： 用于在高速交换环境中进行组播的系统包括逻辑上划分为多个块的数据存储器，多个输出端口和中心代理。 中央代理可以确定存储在一个或多个块中的分组的所有一个或多个输出端口是否已经读取分组。 当分组的一个或多个输出端口中的最后一个从块读取时，可以将分组的每个块添加到可用块池中。

    Purge mechanism in link aggregation group management
    9.
    发明申请
    Purge mechanism in link aggregation group management 有权
    链路聚合组管理中的清除机制

    公开(公告)号：US20060227711A1

    公开(公告)日：2006-10-12

    申请号：US11394892

    申请日：2006-03-31

    申请人： Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： H04J1/16 ,  H04L12/56

    CPC分类号： H04L41/0896 ,  H04L47/125 ,  H04L69/14 ,  Y02D50/30

    摘要： To enable quick movement of communications among links in a link aggregation group, network element use a purge mechanism. A network element implementing the purge mechanism may disable distribution of additional frames to output queues associated with aggregated ports and potentially drop some or all frames from the output queues associated with aggregated ports. In conjunction with the dropping of frames, the network element may exchange one or more marker messages and marker responses with a remote network element. After receiving appropriate responses, the network element may restart distribution of frames to the affected ports.

    摘要翻译： 为了使链路聚合组中链路之间的通信能够快速移动，网元使用清除机制。 实现清除机制的网络元件可以禁止将附加帧分配到与聚合端口相关联的输出队列，并且可能从与聚合端口相关联的输出队列中丢弃一些或所有帧。 结合帧的丢弃，网络元件可以与远程网络元件交换一个或多个标记消息和标记响应。 在收到适当的响应后，网元可能会重新分配帧到受影响的端口。

    Special marker message for link aggregation marker protocol
    10.
    发明授权
    Special marker message for link aggregation marker protocol 有权
    链接聚合标记协议的特殊标记消息

    公开(公告)号：US08451713B2

    公开(公告)日：2013-05-28

    申请号：US11394960

    申请日：2006-03-31

    申请人： Yukihiro Nakagawa ,  Takeshi Shimizu

    发明人： Yukihiro Nakagawa ,  Takeshi Shimizu

    IPC分类号： G01R11/00 ,  G06F15/00 ,  G08C15/00 ,  H04J1/16 ,  H04J3/14 ,  H04L1/00 ,  H04L12/26 ,  H04L12/28

    CPC分类号： H04L12/26 ,  H04L41/0681 ,  H04L43/00 ,  H04L43/0817 ,  H04L45/00 ,  H04L45/22 ,  H04L45/245 ,  H04L45/28 ,  Y02D50/30

    摘要： A network element aggregates a number of its ports together to form a link aggregation group. To move communications from one of the aggregated ports in case that port fails or otherwise becomes inactive, the network element may use exchange a special marker message and response with a remote network element coupled to the failed port. The network elements use proprietary signaling within the special marker message and response to enable the exchange of these messages on an active one of the aggregated ports while identifying the failed one of the ports.

    摘要翻译： 网元将多个端口聚合在一起形成链路聚合组。 为了在端口故障或其他方式变得不活动的情况下从聚合端口之一移动通信，网络元件可以使用与故障端口耦合的远程网络元件交换特殊标记消息和响应。 网络元件在特殊标记消息和响应中使用专有信令，以便在标识端口中的一个端口时，在活动的聚合端口之间交换这些消息。