公开(公告)号：US07633864B2

公开(公告)日：2009-12-15

申请号：US11642427

申请日：2006-12-20

申请人： Darrin P. Johnson ,  Darren J. Reed ,  Erik Nordmark

发明人： Darrin P. Johnson ,  Darren J. Reed ,  Erik Nordmark

IPC分类号： H04L12/26 ,  H04L12/28 ,  H04L12/56 ,  H04J3/16 ,  G06F15/16 ,  G06F15/173

CPC分类号： H04L45/00 ,  H04L63/0209 ,  H04L63/0236 ,  H04L63/0272

摘要： A method for routing a packet. The method includes receiving the packet from a first network into a network interface card (NIC), where the NIC is operatively connected to a host and the host includes a first virtual network stack and a second virtual network stack. The method further includes sending the packet to a first virtual network stack, where the first virtual network stack includes a first filter, a first network layer, and a first transport layer. In addition, the first filter, the first network layer, and the first transport layer are isolated from the second virtual network stack. If the packet is permitted through the first filter in the first virtual network stack, then the packet is sent to a first virtual NIC.

摘要翻译： 一种用于路由数据包的方法。 该方法包括从第一网络接收分组到网络接口卡（NIC），其中NIC可操作地连接到主机，并且主机包括第一虚拟网络堆栈和第二虚拟网络堆栈。 该方法还包括将分组发送到第一虚拟网络栈，其中第一虚拟网络栈包括第一过滤器，第一网络层和第一传输层。 此外，第一过滤器，第一网络层和第一传输层与第二虚拟网络堆栈隔离。 如果通过第一虚拟网络堆栈中的第一过滤器允许数据包，则将数据包发送到第一虚拟NIC。

公开(公告)号：US20080155676A1

公开(公告)日：2008-06-26

申请号：US11642427

申请日：2006-12-20

申请人： Darrin P. Johnson ,  Darren J. Reed ,  Erik Nordmark

发明人： Darrin P. Johnson ,  Darren J. Reed ,  Erik Nordmark

IPC分类号： H04L9/00 ,  H04L12/56

CPC分类号： H04L45/00 ,  H04L63/0209 ,  H04L63/0236 ,  H04L63/0272

摘要： A method for routing a packet. The method includes receiving the packet from a first network into a network interface card (NIC), where the NIC is operatively connected to a host and the host includes a first virtual network stack and a second virtual network stack. The method further includes sending the packet to a first virtual network stack, where the first virtual network stack includes a first filter, a first network layer, and a first transport layer. In addition, the first filter, the first network layer, and the first transport layer are isolated from the second virtual network stack. If the packet is permitted through the first filter in the first virtual network stack, then the packet is sent to a first virtual NIC.

摘要翻译： 一种用于路由数据包的方法。 该方法包括从第一网络接收分组到网络接口卡（NIC），其中NIC可操作地连接到主机，并且主机包括第一虚拟网络堆栈和第二虚拟网络堆栈。 该方法还包括将分组发送到第一虚拟网络栈，其中第一虚拟网络栈包括第一过滤器，第一网络层和第一传输层。 此外，第一过滤器，第一网络层和第一传输层与第二虚拟网络堆栈隔离。 如果通过第一虚拟网络堆栈中的第一过滤器允许数据包，则将数据包发送到第一虚拟NIC。

公开(公告)号：US20130294221A1

公开(公告)日：2013-11-07

申请号：US13465218

申请日：2012-05-07

申请人： Ramkumar Parameswaran ,  Varun Shah ,  Erik Nordmark ,  Ayan Banerjee

发明人： Ramkumar Parameswaran ,  Varun Shah ,  Erik Nordmark ,  Ayan Banerjee

IPC分类号： H04L12/56 ,  G06F11/00

CPC分类号： H04L45/02 ,  H04L12/4641 ,  H04L12/4675 ,  H04L45/026 ,  H04L45/66

摘要： Systems and methods may be provided embodying an optimized Trill LAN network hello mode. The optimized hello mode may allow the number of LAN hellos exchanged to be reduced significantly in a steady state mode of operation. No modifications to the current Trill specification are needed and in a converged state (when designated RBridge election and appointed forwarder appointments are complete), only 1 hello PDU per RBridge is originated in every hello interval.

摘要翻译： 可以提供系统和方法，体现优化的Trill LAN网络问候模式。 优化的hello模式可以允许在稳定状态操作模式下交换的LAN hello的数量显着减少。 不需要修改当前的Trill规范，并且在融合状态（当指定RBridge选举和指定的转发器约会完成时），每个Hello报文间隔只会发送一个每个RBridge的1个hello PDU。

公开(公告)号：US08254383B2

公开(公告)日：2012-08-28

申请号：US12242373

申请日：2008-09-30

申请人： Erik Nordmark

发明人： Erik Nordmark

IPC分类号： H04L12/28 ,  H04L12/56

CPC分类号： H04L45/00 ,  H04L45/54

摘要： The invention relates to a method for routing packets. The method includes receiving a first packet to route, obtaining a first route entry using a first IP address associated with the first packet, where the first route entry includes a first prefix, a first route nexthop generation (RNG) and a first route nexthop pointer (RNP) to a second route entry. The method further includes comparing a route generation (RG) for the second route entry with the first RNG where the RG for the second route entry and the first RNG are equal, obtaining a first neighbor cache entry (NCE) from a first NCE pointer (NP) stored in the first route entry, and routing the first packet using the ARP information in the first NCE, where a route dependency list pointer (RDLP) in the second route entry points to the first route entry.

摘要翻译： 本发明涉及一种用于路由分组的方法。 该方法包括接收第一分组以路由，使用与第一分组相关联的第一IP地址获得第一路由条目，其中第一路由条目包括第一前缀，第一路由下一代生成（RNG）和第一路由下一跳指针 （RNP）到第二路由条目。 该方法还包括将第二路由条目的路由生成（RG）与第二路由条目的RG和第一RNG相等的第一RNG进行比较，从第一NCE指针获得第一相邻高速缓存条目（NCE） NP）存储在第一路由条目中，并且使用第一NCE中的ARP信息来路由第一分组，其中第二路由入口中的路由依赖列表指针（RDLP）指向第一路由条目。

公开(公告)号：US07647436B1

公开(公告)日：2010-01-12

申请号：US11118004

申请日：2005-04-29

申请人： Roland Westrelin ,  Erik Nordmark ,  Nicolas Fugier ,  Eric Lemoine

发明人： Roland Westrelin ,  Erik Nordmark ,  Nicolas Fugier ,  Eric Lemoine

IPC分类号： G06F13/00 ,  G06F3/00 ,  G06F13/28 ,  G06F15/16 ,  G06F11/00 ,  G01R31/08 ,  G08C15/00 ,  H04J1/16 ,  H04J3/14 ,  H04L1/00 ,  H04L12/26

CPC分类号： G01R31/08 ,  G06F3/00 ,  G06F11/00 ,  G06F13/00 ,  G06F13/28 ,  G06F15/16 ,  G08C15/00 ,  H04J1/16 ,  H04J3/14 ,  H04L1/00 ,  H04L1/1874 ,  H04L43/00

摘要： A system that includes a host including at least one per-connection data structure and at least one per-processor data structure, wherein the at least one per-connection data structure is associated with a connection, and an offload engine operatively connected to the host. The engine includes offload engine connection registers and functionality to update the at least one per-connection data structures in the host, wherein the offload engine is configured to send and receive network data on the connection, wherein the host and the offload engine communicate using the at least one per-processor data structure, and wherein the offload engine communicates a status of the connection to the host using the offload engine connection registers.

摘要翻译： 一种包括主机的系统，所述主机包括至少一个每连接数据结构和至少一个每处理器数据结构，其中所述至少一个每连接数据结构与连接相关联，并且卸载引擎可操作地连接到所述主机 。 引擎包括卸载引擎连接寄存器和用于更新主机中的至少一个每连接数据结构的功能，其中卸载引擎被配置为在连接上发送和接收网络数据，其中主机和卸载引擎使用 至少一个每处理器数据结构，并且其中所述卸载引擎使用卸载引擎连接寄存器来传送与所述主机的连接的状态。

公开(公告)号：US20090046601A1

公开(公告)日：2009-02-19

申请号：US12047097

申请日：2008-03-12

申请人： Erik Nordmark ,  Samita Chakrabarti

发明人： Erik Nordmark ,  Samita Chakrabarti

IPC分类号： H04L12/28

CPC分类号： H04W40/08 ,  H04L45/26 ,  H04W40/246 ,  Y02D70/22 ,  Y02D70/324

摘要： In a particular embodiment of the present invention, a particular node in an ad hoc network is configured with a desirable number of neighbor nodes and equipped with a transmission power level hierarchy. The particular node may adjust a transmission power level to a different level in the transmission power level hierarchy, based on a number of responses received from neighbor nodes after the particular node broadcasts a request at current transmission power level. This adjustment process is repeated until the particular node acquires a desirable number of neighbor nodes.

摘要翻译： 在本发明的特定实施例中，自组织网络中的特定节点被配置有期望数量的邻居节点，并配备有发射功率电平层级。 基于在特定节点以当前发射功率电平广播请求之后从邻居节点接收的响应的数量，特定节点可以将发射功率电平调整到发射功率电平层级中的不同电平。 重复该调整处理，直到特定节点获取期望数量的邻居节点。

公开(公告)号：US06578036B1

公开(公告)日：2003-06-10

申请号：US09571398

申请日：2000-05-15

申请人： Jarrett J. Lu ,  Nicolai Kosche ,  Erik Nordmark ,  Jeff Bonwick

发明人： Jarrett J. Lu ,  Nicolai Kosche ,  Erik Nordmark ,  Jeff Bonwick

IPC分类号： G06F1730

CPC分类号： G06F17/30067 ,  Y10S707/99945

摘要： An improved mechanism for efficiently polling file descriptors in a large scale computer system is disclosed. The polling of the file descriptors is performed by first determining, based upon a set of indication information, which file descriptors are eligible for polling. A file descriptor is eligible for polling if the indication information for that file descriptor indicates that the file associated with that file descriptor might (but does not necessarily) have an event pending. Only if a file descriptor is determined to be eligible will the file descriptor and the file associated with the file descriptor be polled. By polling only the eligible file descriptors in a system, the overhead associated with the polling operation is kept to a minimum. As a result, even in a large scale system with a large number of file descriptors, polling is still performed efficiently.

摘要翻译： 公开了一种用于在大型计算机系统中有效地轮询文件描述符的改进机制。 通过首先基于一组指示信息确定哪些文件描述符有资格进行轮询来执行对文件描述符的轮询。 如果该文件描述符的指示信息指示与该文件描述符相关联的文件可能（但不一定））具有事件挂起，则文件描述符可以轮询。 只有当文件描述符被确定为符合条件时，才会轮询文件描述符和文件描述符关联的文件。 通过仅轮询系统中的合格文件描述符，与轮询操作相关联的开销将保持最小。 因此，即使在具有大量文件描述符的大规模系统中，仍然有效地执行轮询。

公开(公告)号：US08659995B2

公开(公告)日：2014-02-25

申请号：US12944271

申请日：2010-11-11

申请人： Sowmini Varadhan ,  Erik Nordmark

发明人： Sowmini Varadhan ,  Erik Nordmark

IPC分类号： G01R31/08

CPC分类号： H04L43/10

摘要： A system (including a processor, a host having a network multipath daemon, and multiple network interfaces) and method involving issuing, by an active interface, an Internet Protocol probe packet (IPPP) to a probe target, issuing, by an alternate interface, a transitive probe packet (TPP) to the active interface, where the alternate interface and the active interface belong to a broadcast domain. The system and method further involve awaiting, by the active interface, a responsive IPPP from the probe target, awaiting by the alternate interface, a responsive TPP from the active interface, and determining network conditions based upon receiving at least one of the responsive IPPP and the responsive TPP.

摘要翻译： 一种系统（包括处理器，具有网络多路径守护进程的主机和多个网络接口）以及涉及通过主动接口向探测目标发布互联网协议探测分组（IPPP）的方法， 传递探测分组（TPP）到活动接口，其中备用接口和活动接口属于广播域。 所述系统和方法还包括通过主动接口等待来自探测目标的响应IPPP，由备用接口等待来自主动接口的响应TPP，以及基于接收到响应IPPP中的至少一个来确定网络条件，以及 反应性TPP。

公开(公告)号：US08180352B2

公开(公告)日：2012-05-15

申请号：US12047097

申请日：2008-03-12

申请人： Erik Nordmark ,  Samita Chakrabarti

发明人： Erik Nordmark ,  Samita Chakrabarti

IPC分类号： H04W72/00

CPC分类号： H04W40/08 ,  H04L45/26 ,  H04W40/246 ,  Y02D70/22 ,  Y02D70/324

摘要： In a particular embodiment of the present invention, a particular node in an ad hoc network is configured with a desirable number of neighbor nodes and equipped with a transmission power level hierarchy. The particular node may adjust a transmission power level to a different level in the transmission power level hierarchy, based on a number of responses received from neighbor nodes after the particular node broadcasts a request at current transmission power level. This adjustment process is repeated until the particular node acquires a desirable number of neighbor nodes.

摘要翻译： 在本发明的特定实施例中，自组织网络中的特定节点被配置有期望数量的邻居节点，并配备有发射功率电平层级。 基于在特定节点以当前发射功率电平广播请求之后从邻居节点接收的响应的数量，特定节点可以将发射功率电平调整到发射功率电平层级中的不同电平。 重复该调整处理，直到特定节点获取期望数量的邻居节点。

公开(公告)号：US07944899B2

公开(公告)日：2011-05-17

申请号：US12058509

申请日：2008-03-28

申请人： Erik Nordmark ,  Samita Chakrabarti

发明人： Erik Nordmark ,  Samita Chakrabarti

IPC分类号： H04W4/00

CPC分类号： H04W40/18 ,  H04W40/08 ,  H04W40/10 ,  Y02D70/164 ,  Y02D70/22 ,  Y02D70/30 ,  Y02D70/324 ,  Y02D70/326

摘要： In a particular embodiment of the present invention, a forwarding node in an ad hoc network collects sets of node information pertaining to its neighbor nodes. An example of a type of node information is history information that indicates previous time periods during which a neighbor node has been available for communication. Other types of node information may include proximity information that indicates how far a neighbor node is from the forwarding node. In selecting a routing path, the forwarding node takes into account information collected in the sets of node information. A routing path can be selected to include a neighbor node having the highest probability to be available for communication. Additionally or alternatively, a routing path may be selected to minimize the power transmission level required for the forwarding node to forward data packets.

摘要翻译： 在本发明的特定实施例中，自组织网络中的转发节点收集与其相邻节点有关的节点信息集合。 一种节点信息的示例是指示相邻节点已经可用于通信的先前时间段的历史信息。 其他类型的节点信息可以包括指示邻居节点与转发节点有多远的邻近信息。 在选择路由路径时，转发节点考虑在节点信息集中收集的信息。 可以选择路由路径以包括具有可用于通信的最高概率的邻居节点。 附加地或替代地，可以选择路由路径以最小化转发节点转发数据分组所需的功率传输级别。