Universal network interface controller
    91.
    发明授权
    Universal network interface controller 有权
    通用网络接口控制器

    公开(公告)号:US09042383B2

    公开(公告)日:2015-05-26

    申请号:US13173189

    申请日:2011-06-30

    CPC classification number: H04L49/358 H04L49/505

    Abstract: A universal network interface controller (UNIC) is provided for interfacing a host computer to a switch fabric, a packet network, or both. The UNIC includes ingress transmit logic designed to transmit switch fabric data in memory associated with the host computer to a switch fabric. The UNIC further includes egress receive logic designed to receive switch fabric data from the switch fabric to store the received switch fabric data in the host memory associated with the host computer. As an option, the ingress transmit logic may be further designed to transmit packet network data in memory associated with the host computer to a packet network, such as Ethernet, and the egress receive logic may be further designed to receive the packet network data from the packet network and to store the received switch fabric data and the received packet network data in the host memory associated with the host computer.

    Abstract translation: 提供通用网络接口控制器(UNIC),用于将主机与交换结构,分组网络或两者接口。 UNIC包括入口传输逻辑,用于将与主计算机相关联的存储器中的交换结构数据传送到交换结构。 UNIC还包括出口接收逻辑,用于从交换结构接收交换结构数据,以将接收的交换结构数据存储在与主计算机相关联的主机存储器中。 作为选择,入口传输逻辑可以进一步被设计为将与主计算机相关联的存储器中的分组网络数据传送到诸如以太网的分组网络,并且出口接收逻辑可以被进一步设计为从 并将接收到的交换结构数据和接收到的分组网络数据存储在与主计算机相关联的主机存储器中。

    Apparatus and method for storing a port number in association with one or more addresses
    92.
    发明授权
    Apparatus and method for storing a port number in association with one or more addresses 有权
    用于存储与一个或多个地址相关联的端口号的装置和方法

    公开(公告)号:US08984160B2

    公开(公告)日:2015-03-17

    申请号:US13295476

    申请日:2011-11-14

    CPC classification number: H04L49/358 H04L49/25 H04L49/557 H04L49/65

    Abstract: A port number is stored in a memory in association with one or more addresses using a data block by which at least one port number associated with a predetermined number of consecutive addresses is written into the memory. Further, a first port number assigned to output ports each being provided for different one of a plurality of relay nodes is stored in association with first consecutive addresses in such a manner that one or more data blocks including the first port number associated with the first consecutive addresses are written into the memory.

    Abstract translation: 使用数据块将端口号与一个或多个地址相关联地存储在存储器中,通过该数据块将与预定数量的连续地址相关联的至少一个端口号写入存储器。 此外,分配给多个中继节点中的不同一个中继节点的输出端口的第一端口号与第一连续地址相关联地存储,使得一个或多个数据块包括与第一连续的相关联的第一端口号 地址被写入存储器。

    SYSTEM AND METHOD FOR SUPPORTING MULTI-HOMED FAT-TREE ROUTING IN A MIDDLEWARE MACHINE ENVIRONMENT
    94.
    发明申请
    SYSTEM AND METHOD FOR SUPPORTING MULTI-HOMED FAT-TREE ROUTING IN A MIDDLEWARE MACHINE ENVIRONMENT 有权
    用于在中间件机器环境中支持多重FAT-TREE路由的系统和方法

    公开(公告)号:US20150030034A1

    公开(公告)日:2015-01-29

    申请号:US14226288

    申请日:2014-03-26

    CPC classification number: H04L49/25 H04L45/128 H04L45/24 H04L49/15 H04L49/358

    Abstract: A system and method can support multi-homed routing in a network environment, which can be based on InfiniBand architecture using a fat-tree or a similar topology. The system can provide an end node that is associated with a switch port on a leaf switch in a network fabric. Then, the system can perform routing for each of a plurality of ports on the end node, and ensure that the plurality of ports on the end node take mutually independent paths.

    Abstract translation: 系统和方法可以支持网络环境中的多宿主路由,这可以基于使用胖树或类似拓扑的InfiniBand架构。 系统可以提供与网络结构中的叶片交换机上的交换机端口相关联的端节点。 然后,系统可以为终端节点上的多个端口中的每个端口执行路由,并确保端节点上的多个端口采用相互独立的路径。

    Packet switching based on global identifier
    95.
    发明授权
    Packet switching based on global identifier 有权
    基于全局标识符的分组交换

    公开(公告)号:US08842671B2

    公开(公告)日:2014-09-23

    申请号:US13154458

    申请日:2011-06-07

    CPC classification number: H04L49/602 H04L49/358

    Abstract: A communication method in a network operating in accordance with a standard that allocates a given number of bits m for layer-2 addressing of nodes in the network. The method includes accepting at a layer-2 switch in the network an assignment to one or more nodes in the network of respective layer-2 extended addresses, each including n=m+k bits, k>0. A given data packet is received at the switch for forwarding. The given data packet includes a layer-2 destination address and a layer-3 destination address in accordance with the standard. The layer-3 destination address includes t bits, t≧k. The given data packet is forwarded from the switch to one of the nodes by reading from the given data packet and combining the layer-2 destination address and k bits from the layer-3 destination address so as to reconstruct the n bits of the extended layer-2 address of the one of the nodes.

    Abstract translation: 根据标准操作的网络中的通信方法,所述标准为网络中的节点的第2层寻址分配给定数量的位m。 该方法包括在网络中的第二层交换机处接收对相应的第二层扩展地址的网络中的一个或多个节点的分配,每个节点包括n = m + k个比特,k> 0。 在交换机处接收给定的数据包以进行转发。 给定的数据分组包括根据标准的二层目的地址和三层目的地址。 第3层目的地址包括t位,t≥k。 通过从给定的数据包中读取并将第二层目的地址和第三层目的地址的k位组合起来,将给定的数据包从交换机转发到其中一个节点,以重构扩展层的n位 -2地址之一的节点。

    Systems and Methods for Paravirtualized IP Over Infiniband Bridging
    96.
    发明申请
    Systems and Methods for Paravirtualized IP Over Infiniband Bridging 有权
    非虚拟化IP跨无线电桥接的系统和方法

    公开(公告)号:US20140241355A1

    公开(公告)日:2014-08-28

    申请号:US13777891

    申请日:2013-02-26

    Inventor: Michael Tsirkin

    CPC classification number: H04L45/72 H04L49/358 H04L49/70

    Abstract: Systems and methods for paravirtualized Internet Protocol over Infiniband (IPoIB) bridging for virtual machines. An example method may comprise associating, by a hypervisor executing on a host computer system, an Infiniband (IB) interface of the host computer system with a paravirtualized IPoIB interface of a virtual machine executing on the host computer system; receiving, via the paravirtualized IPoIB interface, a paravirtualized IPoIB packet originated by the virtual machine, the paravirtualized IPoIB packet comprising an IP packet and a first header including an IPoIB hardware address; translating the paravirtualized IPoIB packet into an IPoIB frame comprising the IP packet and a second header including the IPoIB hardware address; and forwarding the IPoIB frame via the IB interface to one of: an IB switch, an external IB node.

    Abstract translation: 针对虚拟机的Infiniband(IPoIB)桥接半虚拟化互联网协议的系统和方法。 示例性方法可以包括通过在主计算机系统上执行的管理程序将主计算机系统的Infiniband(IB)接口与在主计算机系统上执行的虚拟机的半虚拟化IPoIB接口相关联; 通过所述半虚拟IPoIB接口接收由所述虚拟机发起的半虚拟化IPoIB分组,所述半虚拟化IPoIB分组包括IP分组和包括IPoIB硬件地址的第一头部; 将所述半虚拟化IPoIB分组转换成包括所述IP分组的IPoIB帧以及包括所述IPoIB硬件地址的第二报头; 并通过IB接口将IPoIB帧转发到IB交换机,外部IB节点之一。

    Distributed chassis architecture having integrated service appliances
    97.
    发明授权
    Distributed chassis architecture having integrated service appliances 有权
    具有集成服务设备的分布式机箱结构

    公开(公告)号:US08773999B2

    公开(公告)日:2014-07-08

    申请号:US13281481

    申请日:2011-10-26

    CPC classification number: H04L49/358 H04L49/355

    Abstract: A distributed virtual chassis comprises scaled-out fabric coupler (SFC) boxes. Each SFC box has fabric ports and a cell-based switch fabric for switching cells associated with a packet among the SFC fabric ports of that SFC box. Distributed line cards (DLCs) include switching DLCs and an appliance DLC (A-DLC). Each switching DLC has network ports. Each switching DLC and A-DLC has DLC fabric ports. Each switching DLC and A-DLC is connected to each of the SFC boxes. The A-DLC provides an upper layer service for packets arriving on the network ports of the switching DLCs. To forward a packet to the A-DLC, a switching DLC divides the packet into cells and distributes the cells among the SFC boxes. The SFC boxes forward the cells to the A-DLC, and the A-DLC reassembles the packet from the cells and provides the upper layer service to the packet.

    Abstract translation: 分布式虚拟机箱包括扩展结构耦合器(SFC)框。 每个SFC框具有Fabric端口和一个基于单元的交换结构,用于在SFC框的SFC结构端口之间切换与数据包相关联的单元。 分布式线路卡(DLC)包括切换DLC和设备DLC(A-DLC)。 每个交换DLC都有网络端口。 每个交换DLC和A-DLC都有DLC接口。 每个切换DLC和A-DLC连接到每个SFC框。 A-DLC为到达切换DLC的网络端口的分组提供了上层服务。 要将数据包转发到A-DLC,切换DLC将数据包划分为单元格,并在SFC框之间分配单元。 SFC盒将单元转发到A-DLC,A-DLC重新组合来自单元的数据包,并为数据包提供上层服务。

    Shared memory message switch and cache
    98.
    发明授权
    Shared memory message switch and cache 有权
    共享内存消息切换和缓存

    公开(公告)号:US08719456B2

    公开(公告)日:2014-05-06

    申请号:US12986099

    申请日:2011-01-06

    CPC classification number: H04L49/254 G06F13/4059 H04L49/35 H04L49/358

    Abstract: A method and apparatus are described to provide shared switch and cache memory. The apparatus may comprise a message switch module, a cache controller module, and shared switch and cache memory to provide shared memory to the message switch module and to the cache controller module. The cache controller module may comprise pointer memory to store a plurality of pointers, each pointer pointing to a location in the shared switch and cache memory (e.g., point to a message header partition in the shared switch and cache memory). If there is a corresponding pointer, a memory read response may be sent to the requesting agent. If there is no corresponding pointer, a write data request may be sent to a corresponding destination agent and, in response to receiving the requested data, a pointer to the stored data in the pointer memory may be provided.

    Abstract translation: 描述了一种提供共享交换机和高速缓冲存储器的方法和装置。 该装置可以包括消息交换模块,高速缓存控制器模块以及共享交换机和高速缓冲存储器,以向消息交换模块和高速缓存控制器模块提供共享存储器。 高速缓存控制器模块可以包括指针存储器以存储多个指针,每个指针指向共享交换机和高速缓冲存储器中的位置(例如,指向共享交换机和高速缓冲存储器中的消息头部分区)。 如果存在对应的指针,则可以向请求代理发送存储器读取响应。 如果没有对应的指针,则可以向对应的目的地代理发送写数据请求,并且响应于接收到所请求的数据,可以提供指向存储在指针存储器中的存储数据的指针。

Patent Agency Ranking