公开(公告)号：US20080140358A1

公开(公告)日：2008-06-12

申请号：US11635404

申请日：2006-12-06

Applicant: Kevin J. Lang ,  Vivek B. Tawde

Inventor： Kevin J. Lang ,  Vivek B. Tawde

IPC: G06F17/50 ,  G06F17/00 ,  G06F15/00 ,  G06F3/048

CPC classification number: G06F17/504 ,  G06F17/509 ,  G06F2217/04

Abstract: A system and method to identify target node graphs from predetermined seed node subsets are described. A one-sided constrained spectral embedding of a graph using a predetermined set of seed nodes within the graph is computed. Next, a threshold boundary is identified to define a target area for a target subgraph within the embedded graph using predetermined network-flow calculations, the target subgraph including the predetermined set of seed nodes. Finally, the target subgraph is identified based on the calculated threshold boundary.

Abstract translation: 描述了从预定种子节点子集识别目标节点图的系统和方法。 计算图中使用图中预定的种子节点集合的单边约束频谱嵌入。 接下来，识别阈值边界，以使用预定的网络流计算来定义嵌入图中的目标子图的目标区域，目标子图包括预定的种子节点集合。 最后，基于计算的阈值边界来识别目标子图。

公开(公告)号：US06826564B2

公开(公告)日：2004-11-30

申请号：US09900704

申请日：2001-07-06

Applicant: Nicholas Thompson ,  William Welch ,  Kevin J. Lang ,  Steven McCanne ,  Matthew G. Podolsky

Inventor： Nicholas Thompson ,  William Welch ,  Kevin J. Lang ,  Steven McCanne ,  Matthew G. Podolsky

IPC: G06F1730

CPC classification number: H04L41/046 ,  H04L41/0213 ,  H04L41/04 ,  Y10S707/99934 ,  Y10S707/99935

Abstract: In a network, a query originator injects queries of network devices into the network at a query node using query messages. The network transports the query messages to the network devices, or to network nodes at which queries about the network devices can be answered. Query responses from the network devices or network nodes are directed through the network to a collection node or nodes. As an internal network node receives multiple query responses from network devices, the internal network node might aggregate, as needed, the multiple query responses into an aggregated query response that preferably occupies less bandwidth than the aggregated multiple query responses. Where the result desired at the collection node is a computed function of the multiple query responses, the computed function can be performed at each internal network node on the multiple query responses received at that node, thus distributing the computation needed to form a collected response to the query. Queries might request real-time or non-real-time responses and queries might request one response, periodic responses or one response for each occurrence of an event. The internal network nodes might store lookup tables, or access a centralized lookup table, that specify details of queries, thus obviating the need for each query message to carry all of the necessary details of the query.

Abstract translation: 在网络中，查询发起者使用查询消息将网络设备的查询注入到查询节点的网络中。 网络将查询消息传输到网络设备，或网络节点，可以应答有关网络设备的查询。 来自网络设备或网络节点的查询响应通过网络引导到收集节点。 当内部网络节点从网络设备接收多个查询响应时，内部网络节点可以根据需要将多个查询响应聚合成聚合查询响应，其优选地占据比聚合的多个查询响应更少的带宽。 在收集节点所期望的结果是多个查询响应的计算函数的情况下，可以在该节点处接收的多个查询响应的每个内部网络节点处执行所计算的函数，从而将形成收集的响应所需的计算分配到 查询。 查询可能会请求实时或非实时响应，查询可能会请求一个响应，定期响应或每次事件发生的一个响应。 内部网络节点可能存储查找表，或访问指定查询详细信息的集中式查找表，从而避免了每个查询消息需要携带查询的所有必要细节。

公开(公告)号：US07805667B2

公开(公告)日：2010-09-28

申请号：US11635404

申请日：2006-12-06

Applicant: Kevin J. Lang ,  Vivek B. Tawde

Inventor： Kevin J. Lang ,  Vivek B. Tawde

IPC: G06F17/00

CPC classification number: G06F17/504 ,  G06F17/509 ,  G06F2217/04

Abstract: A system and method to identify target node graphs from predetermined seed node subsets are described. A one-sided constrained spectral embedding of a graph using a predetermined set of seed nodes within the graph is computed. Next, a threshold boundary is identified to define a target area for a target subgraph within the embedded graph using predetermined network-flow calculations, the target subgraph including the predetermined set of seed nodes. Finally, the target subgraph is identified based on the calculated threshold boundary.

Abstract translation: 描述了从预定种子节点子集识别目标节点图的系统和方法。 计算图中使用图中预定的种子节点集合的单边约束频谱嵌入。 接下来，识别阈值边界，以使用预定的网络流计算来定义嵌入图中的目标子图的目标区域，目标子图包括预定的种子节点集合。 最后，基于计算的阈值边界来识别目标子图。