公开(公告)号：US09552550B2

公开(公告)日：2017-01-24

申请号：US14276363

申请日：2014-05-13

申请人： Cisco Technology, Inc.

发明人： Jean-Philippe Vasseur ,  Sukrit Dasgupta

IPC分类号： G06F15/18 ,  G06N5/04 ,  H04L12/26 ,  H04L12/24 ,  G06N99/00 ,  H04L12/751 ,  H04L12/815

CPC分类号： G06N5/04 ,  G06N99/005 ,  H04L41/5003 ,  H04L43/062 ,  H04L43/0876 ,  H04L45/08 ,  H04L47/22

摘要： In one embodiment, a committed information rate (CIR) prediction is received from a machine learning model that corresponds to a predicted average traffic rate supported by a network connection. A traffic shaping strategy is adjusted based on the CIR prediction. A rate at which data is communicated over the network connection may be based on the traffic shaping policy. The effects of the adjusted traffic shaping strategy are also monitored. Feedback is further provided to the machine learning model based on the monitored effects of the adjusted traffic shaping strategy.

摘要翻译： 在一个实施例中，从与由网络连接支持的预测平均业务速率相对应的机器学习模型接收到承诺信息速率（CIR）预测。 根据CIR预测调整流量整形策略。 通过网络连接传送数据的速率可以基于流量整形策略。 调整流量整形策略的效果也受到监控。 基于调整后的流量整形策略的监控效果，进一步向机器学习模型提供反馈。

公开(公告)号：US09507745B1

公开(公告)日：2016-11-29

申请号：US14594299

申请日：2015-01-12

申请人： Ian Rudolf Bratt ,  Carl G. Ramey ,  Matthew Mattina

发明人： Ian Rudolf Bratt ,  Carl G. Ramey ,  Matthew Mattina

IPC分类号： G06F15/16 ,  G06F13/40 ,  G06F13/42

CPC分类号： G06F9/5022 ,  G06F9/4825 ,  G06F9/4881 ,  G06F9/544 ,  G06F11/1482 ,  G06F12/00 ,  G06F12/02 ,  G06F12/0806 ,  G06F12/0808 ,  G06F12/0844 ,  G06F12/0848 ,  G06F12/0864 ,  G06F12/0891 ,  G06F12/10 ,  G06F12/1027 ,  G06F12/12 ,  G06F12/121 ,  G06F12/123 ,  G06F13/4068 ,  G06F13/4221 ,  G06F15/16 ,  G06F15/8023 ,  G06F21/79 ,  G06F2212/502 ,  G06F2212/601 ,  H04L45/08 ,  H04L45/22 ,  H04L45/28 ,  H04L47/823 ,  H04L49/109

摘要： Communicating among cores in a computing system comprising a plurality of cores, each core comprising a processor and a switch, includes: routing a packet from a core or from a device coupled to at least one core to a destination over a route including one or more cores, with an order of dimensions associated with the route being selected dynamically upon construction of the packet; routing the packet to a first core in the route over the first selected dimension; and routing the packet from the first core to the destination over the second dimension.

公开(公告)号：US20160344571A1

公开(公告)日：2016-11-24

申请号：US14720760

申请日：2015-05-23

申请人： Cisco Technology, Inc.

发明人： Grant Allen Marvin ,  Timothy J. Passaro ,  Justin Joel Delegard ,  James Michael Roewe

IPC分类号： H04L12/44 ,  H04W48/16 ,  H04L12/46

CPC分类号： H04L12/4641 ,  H04L12/462 ,  H04L45/08 ,  H04L45/18 ,  H04W48/16

摘要： Various implementations disclosed herein include systems, methods and apparatuses that determine connections between disconnected partial trees. For example, in some implementations, a method includes obtaining a respective device identifier of a first network device included in a first one of a plurality of partial trees included in a network, wherein each partial tree includes at least one network device, and connectivity from the at least one network device to other partial trees is undefined based on associated uplink metadata; querying a plurality of compliant devices within the plurality of partial trees in order to trace a path traversal of the respective device identifier within the network; and selecting one of the plurality of compliant devices as connecting the first network device to a second one of the plurality of partial trees based at least on query responses provided by the plurality of compliant devices.

摘要翻译： 本文公开的各种实现方式包括确定断开的部分树之间的连接的系统，方法和装置。 例如，在一些实现中，一种方法包括获得包括在包括在网络中的多个部分树中的第一个中的第一网络设备的相应设备标识符，其中每个部分树包括至少一个网络设备，以及从 基于相关联的上行链路元数据，对其他部分树的至少一个网络设备是未定义的; 查询多个部分树中的多个兼容设备，以便跟踪网络内相应设备标识符的路径遍历; 以及至少基于由所述多个兼容设备提供的查询响应，选择所述多个兼容设备中的一个将所述第一网络设备连接到所述多个部分树中的第二部分树。

公开(公告)号：US20160330102A1

公开(公告)日：2016-11-10

申请号：US14707740

申请日：2015-05-08

申请人： Arista Networks, Inc.

发明人： William Fenner ,  Prakash Badrinaryananan

IPC分类号： H04L12/755 ,  H04L12/26 ,  H04L12/771

CPC分类号： H04L45/021 ,  H04L45/02 ,  H04L45/08 ,  H04L45/12 ,  H04L45/54 ,  H04L45/56 ,  H04L47/28

摘要： A method for programming route hardware in network devices. The method includes: receiving, by a network device, route updates, selecting from the route updates, a set of selected routes to be used for routing, assigning multiple timestamps to the set of selected routes, programming the set of selected routes into a software-implemented forwarding information base (FIB), segmenting the set of selected routes in the software-implemented FIB into at least a first batch of routes and a second batch of routes, programming the first batch of routes into a hardware-implemented FIB, after the programming, determining a last timestamp, where the last time stamp is the oldest timestamp of the timestamps where all routes associated with the last timestamp have been successfully programmed into the hardware-implemented FIB, and advertising all routes of the first batch that are associated with a timestamp that is no older than the last timestamp.

摘要翻译： 一种在网络设备中编程路由硬件的方法。 该方法包括：由网络设备接收路由更新，从路由更新中选择要用于路由的一组选择的路由，将多个时间戳分配给所选路由的集合，将所选择的路由集合编程成软件 实现的转发信息库（FIB），将软件实现的FIB中的所选路由集合分段为至少第一批路由和第二批路由，将第一批路由编程为硬件实现的FIB，之后 编程，确定最后时间戳，其中最后时间戳是时间戳的最早的时​​间戳，其中与最后时间戳相关联的所有路线已被成功地编程到硬件实现的FIB中，并且广告与第一批相关联的所有路线 时间戳不晚于最后一个时间戳。

公开(公告)号：US20160323150A1

公开(公告)日：2016-11-03

申请号：US15207706

申请日：2016-07-12

申请人： lntel Corporation

发明人： Michael Heinz ,  Todd Rimmer ,  James Kunz ,  Mark Debbage

IPC分类号： H04L12/24 ,  H04L12/753 ,  H04L12/751

CPC分类号： H04L41/12 ,  G06F9/5066 ,  H04L41/0213 ,  H04L41/046 ,  H04L41/0893 ,  H04L45/08 ,  H04L45/48

摘要： System, method, and apparatus for improving the performance of collective operations in High Performance Computing (HPC). Compute nodes in a networked HPC environment form collective groups to perform collective operations. A spanning tree is formed including the compute nodes and switches and links used to interconnect the compute nodes, wherein the spanning tree is configured such that there is only a single route between any pair of nodes in the tree. The compute nodes implement processes for performing the collective operations, which includes exchanging messages between processes executing on other compute nodes, wherein the messages contain indicia identifying collective operations they belong to. Each switch is configured to implement message forwarding operations for its portion of the spanning tree. Each of the nodes in the spanning tree implements a ratcheted cyclical state machine that is used for synchronizing collective operations, along with status messages that are exchanged between nodes. Transaction IDs are also used to detect out-of-order and lost messages.

公开(公告)号：US09456054B2

公开(公告)日：2016-09-27

申请号：US12338175

申请日：2008-12-18

申请人： Van L. Jacobson ,  Diana K. Smetters

发明人： Van L. Jacobson ,  Diana K. Smetters

IPC分类号： G06F12/14 ,  G06F12/16 ,  H04L29/08 ,  G06F15/173 ,  H04L12/751

CPC分类号： H04L63/0227 ,  G06F15/173 ,  H04L9/006 ,  H04L9/14 ,  H04L9/30 ,  H04L9/3247 ,  H04L45/08 ,  H04L45/74 ,  H04L47/20 ,  H04L47/32 ,  H04L63/06 ,  H04L63/123 ,  H04L63/1458 ,  H04L63/20 ,  H04L67/327

摘要： One embodiment of the present invention provides a system for controlling the spread of interests and content in a content centric network (CCN). During operation, the system maintains a routing policy for content data. The system also receives a packet associated with a piece of content or an interest for the content. Next, the system determines that the structured name included in the packet is within the namespace specified in the routing policy. The system further determines that the packet satisfies the condition in the routing policy. Subsequently, the system routes the packet based on in part the action corresponding to the condition as specified in the routing policy.

摘要翻译： 本发明的一个实施例提供了一种用于控制以内容为中心的网络（CCN）的兴趣和内容的扩展的系统。 在操作期间，系统维护内容数据的路由策略。 该系统还接收与一条内容相关联的分组或该内容的兴趣。 接下来，系统确定包中包含的结构名称在路由策略中指定的命名空间内。 系统还确定分组满足路由策略中的条件。 随后，系统部分地基于与路由策略中指定的条件相对应的动作来路由分组。

公开(公告)号：US20160192211A1

公开(公告)日：2016-06-30

申请号：US14621118

申请日：2015-02-12

申请人： Korea Advanced Institute of Science and Technology

发明人： Hong Shik PARK ,  Sang Hyun LEE

IPC分类号： H04W24/02 ,  H04W72/04

CPC分类号： H04W24/02 ,  H04L45/08 ,  H04L45/26 ,  H04L45/38 ,  H04L47/28 ,  H04W28/0252 ,  H04W40/24 ,  H04W84/18

摘要： A cross-layer framework in a wireless mesh network using a bio-inspired algorithm and an operation method thereof are provided. The cross-layer framework includes a data structure configured to be formed in each node of the wireless mesh network and to collect and update information of each node through an ant packet and a cross-layer unit configured to perform at least one or more of channel assignment, routing, link scheduling, buffer management, and frame scheduling with respect to a control data flow of the data structure and to dynamically assign channels.

摘要翻译： 提供了使用生物启发算法的无线网状网络中的跨层框架及其操作方法。 跨层框架包括被配置为形成在无线网状网络的每个节点中的数据结构，并且通过蚂蚁包收集和更新每个节点的信息，并且跨层单元被配置为执行至少一个或多个信道 分配，路由，链路调度，缓冲器管理和帧调度相关的数据结构的控制数据流并且动态分配信道。

公开(公告)号：US20160071014A1

公开(公告)日：2016-03-10

申请号：US14710418

申请日：2015-05-12

申请人： Pivotal Software, Inc.

发明人： Michael Brand ,  Lyndon John Adams ,  David Russell Brown ,  Kee Siong Ng

IPC分类号： G06N5/04 ,  G06N99/00

CPC分类号： G06N20/00 ,  G06N5/02 ,  G06N5/04 ,  H04L45/08

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for stream processing. One method includes receiving an event stream of first events by a first plurality of first local modelers of a stream processing system. Each local modeler processes a portion of received events of the event stream according to a first set of operations, the operations including aggregating information associated with each event to generate first aggregated information. A second plurality of second local modelers similarly generates second aggregated information from an event stream of second events. First and second local modelers provide, to a first central modeler, first and second aggregated information. A set of parameters of a respective machine learning model is determined by the first central modeler using the received aggregated information.

摘要翻译： 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于流处理。 一种方法包括由流处理系统的第一多个第一局部建模器接收第一事件的事件流。 每个本地建模器根据第一组操作处理事件流的接收事件的一部分，所述操作包括聚合与每个事件相关联的信息以生成第一聚合信息。 第二多个第二局部建模器类似地从第二事件的事件流生成第二聚合信息。 第一和第二局部建模者向第一中心建模者提供第一和第二聚合信息。 使用所接收的聚合信息由第一中央建模者确定相应机器学习模型的一组参数。

公开(公告)号：US20160071011A1

公开(公告)日：2016-03-10

申请号：US14710352

申请日：2015-05-12

申请人： Pivotal Software, Inc.

发明人： Michael Brand ,  Lyndon John Adams ,  David Russell Brown ,  Kee Siong Ng

IPC分类号： G06N5/02 ,  G06N99/00 ,  H04L12/751

CPC分类号： G06N99/005 ,  G06N5/02 ,  G06N5/04 ,  H04L45/08

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for routing events of an event stream. One of the methods includes operations of receiving, by a router, events of an event stream; providing each event, by the router, to a respective local modeler selected by the router according to an initial routing strategy, the respective local modeler being selected from multiple local modelers; aggregating, by each local modeler in parallel, information associated with each event received by the local modeler to generate aggregated information; providing, to a central modeler, the aggregated information generated by the one or more local modelers; determining, by the central modeler, parameters of a machine learning model using the aggregated information received by the central modeler and generating an updated routing strategy based on the parameters of the machine learning model; and providing the updated routing strategy to the router.

摘要翻译： 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于路由事件流的事件。 方法之一包括由路由器接收事件流的事件的操作; 根据初始路由策略将由路由器提供的每个事件提供给由路由器选择的相应的本地建模器，相应的本地建模器从多个本地建模器中选择; 由每个本地建模者并行地聚合与由本地建模者接收的每个事件相关联的信息以生成聚合信息; 向中央建模者提供由一个或多个本地建模者生成的聚合信息; 由中央建模者使用由中央建模者接收的聚合信息确定机器学习模型的参数，并且基于机器学习模型的参数生成更新的路由策略; 并向路由器提供更新的路由策略。

公开(公告)号：US09270491B2

公开(公告)日：2016-02-23

申请号：US13511651

申请日：2011-11-14

申请人： Mordehai Margalit ,  Benjamin Maytal

发明人： Mordehai Margalit ,  Benjamin Maytal

IPC分类号： H04L12/64 ,  H04L12/751 ,  H04L12/721

CPC分类号： H04L12/6418 ,  H04L45/08 ,  H04L45/12

摘要： Technologies are generally described for providing a system of scalable networked devices that may be mapped employing a naming convention that enables an optimized route to the devices through device search algorithms for network-connected and non-network connected devices. The naming convention and corresponding algorithms may enable users to search for a desired functionality, access suitable scalable networked device(s), and utilize the desired functionality. Furthermore, a spider algorithm may be employed to continuously map and update a mesh of scalable networked devices.

摘要翻译： 一般来说，技术被描述为提供可扩展联网设备的系统，该系统可以使用命名约定进行映射，该命名约定通过用于网络连接和非网络连接的设备的设备搜索算法实现到设备的优化路由。 命名约定和相应的算法可以使用户能够搜索期望的功能，访问合适的可扩展联网设备，并且利用期望的功能。 此外，可以采用蜘蛛算法来连续地映射和更新可伸缩网络设备的网格。