公开(公告)号：US20130336117A1

公开(公告)日：2013-12-19

申请号：US13524781

申请日：2012-06-15

申请人： Desmond Yan ,  Qun Zheng

发明人： Desmond Yan ,  Qun Zheng

IPC分类号： H04L12/26

CPC分类号： H04J3/0667 ,  H04L45/245 ,  Y02D50/30

摘要： A Precision Timing Protocol (PTP) is implemented over a Link Aggregation Group (LAG) formed by multiple ports of a network node, where PTP traffic goes through the same physical link between the network node and a peer network node on both the transmit and return paths. When the network node receives a PTP message that identifies a PTP stream from the peer network node through a given PTP-LAG port, it declares itself as an active port and the other PTP-LAG ports as standby for the PTP stream. The PTP stream is transmitted from the network node to the peer network node through the active port only, to maintain symmetry of the PTP stream's transmission paths between the network node and the peer network node. The network node processes exchanged messages of the PTP stream to perform timing synchronization with the peer network node.

摘要翻译： 精确定时协议（PTP）通过由网络节点的多个端口形成的链路聚合组（LAG）来实现，其中PTP流量在传输和返回上通过网络节点和对等网络节点之间的相同物理链路 路径。 当网络节点通过给定的PTP-LAG端口接收到来自对等网络节点的PTP流的PTP消息时，它将其自身声明为活动端口，另一个PTP-LAG端口作为PTP流的备用。 PTP流仅通过活动端口从网络节点传输到对等网络节点，以保持PTP流在网络节点和对等网络节点之间的传输路径的对称性。 网络节点处理PTP流的交换消息，以与对等网络节点进行定时同步。

公开(公告)号：US20130070592A1

公开(公告)日：2013-03-21

申请号：US13237745

申请日：2011-09-20

申请人： Minghua Chen ,  Desmond Yan

发明人： Minghua Chen ,  Desmond Yan

IPC分类号： H04L12/26

CPC分类号： H04L47/125 ,  H04L47/10 ,  H04L47/22

摘要： A method and apparatus to shape packet traffic rates of packets in-transit from a plurality of channels to a port, monitor a status of each queue, subtract a CIR from a port load value when an empty queue becomes active, add the CIR to the port load value when an active queue becomes empty, add a difference between an EIR and the CIR for a channel to a port excess value when a queue of the channel becomes excess, subtract the difference for the channel from the port excess value when the excess queue for the channel becomes not excess, calculate an oversubscription ratio based on the port load value and port excess value, calculate a shaping rate for each of the plurality of channels, and shape the traffic for each of the plurality channels using the shaping rate.

摘要翻译： 一种将从多个信道转发到的分组的分组业务速率整形到端口的方法和装置，监视每个队列的状态，当空队列变为活动时从端口负载值中减去CIR，将CIR添加到 端口负载值当活动队列变空时，当通道的队列过多时，EIR和通道的CIR之间的差异会增加到端口超出值，当超出该值时，减去通道与端口超出值的差值 通道的队列变得不够多，基于端口负载值和端口过剩值计算超额预订比率，计算多个信道中的每个信道的整形速率，并且使用整形率对多个信道中的每个信道进行整形。

公开(公告)号：US20130039178A1

公开(公告)日：2013-02-14

申请号：US13209319

申请日：2011-08-12

申请人： Minghua Chen ,  Desmond Yan ,  Blair Mackiewich

发明人： Minghua Chen ,  Desmond Yan ,  Blair Mackiewich

IPC分类号： H04L12/26

CPC分类号： H04L47/22 ,  H04L47/12 ,  H04L47/19 ,  H04L47/20 ,  H04L47/2408 ,  H04L47/2416 ,  H04L47/2441 ,  H04L47/29 ,  H04L47/30 ,  H04L47/39

摘要： Apparatus and techniques for scheduling packets to reduce congestion in a packet data network are described. In one embodiment of the invention, packets are received through each of a plurality of channels and are stored in a respective queue for each channel. The status of each queue is monitored as empty or not empty. When an empty queue becomes not empty, then bandwidth allocation value is added for the respective channel to a port load value. When a not empty queue becomes empty, then the bandwidth allocation for the respective channel is subtracted from the port load value. The traffic rate through the port is then scaled based upon the port load value.

摘要翻译： 描述了用于调度分组以减少分组数据网络中的拥塞的装置和技术。 在本发明的一个实施例中，分组通过多个信道中的每一个被接收，并且被存储在每个信道的相应队列中。 每个队列的状态被监视为空或不为空。 当空队列不为空时，将相应通道的带宽分配值添加到端口负载值。 当不为空的队列变空时，从端口负载值中减去相应通道的带宽分配。 然后通过端口的流量速率根据端口负载值进行缩放。

公开(公告)号：US20130232504A1

公开(公告)日：2013-09-05

申请号：US13412521

申请日：2012-03-05

申请人： Desmond Yan ,  Thomas Ng

发明人： Desmond Yan ,  Thomas Ng

IPC分类号： G06F9/46

CPC分类号： G06F1/3206 ,  G06F1/3287 ,  G06F9/5094 ,  Y02D10/171 ,  Y02D10/22 ,  Y02D50/20

摘要： In one aspect, the present invention reduces average power consumption in a distributed processing system by concentrating an overall processing load to the minimum number of processing units required to maintain a defined level of processing redundancy. When the required number of active processing units is fewer than all available processing units, the inactive processing units may be held in a reduced-power condition. The present invention thereby maintains the defined level of processing redundancy for reallocating jobs responsive to the failure of one of the active processing units, while reducing power consumption and simplifying jobs allocation and re-allocation when expanding or shrinking the active set of processing units responsive to changing processing load. As a non-limiting example, the distributed processing system is implemented within a telecommunications network router or other apparatus having a configured set of processing cards, such as control-plane processing cards.

摘要翻译： 一方面，本发明通过将整体处理负荷集中在维持规定的处理冗余水平所需的最小数量的处理单元来降低分布式处理系统中的平均功耗。 当所需数量的活动处理单元少于所有可用处理单元时，非活动处理单元可以保持在降低功率状态。 因此，本发明可以响应于一个活动处理单元的故障而维持用于重新分配作业的定义的处理冗余级别，同时在扩展或缩小有效的一组处理单元时降低功耗并简化作业分配和重新分配 改变处理负荷。 作为非限制性示例，分布式处理系统在具有配置的处理卡集合（诸如控制平面处理卡）的电信网路由器或其他设备内实现。

公开(公告)号：US08681616B2

公开(公告)日：2014-03-25

申请号：US13209319

申请日：2011-08-12

申请人： Minghua Chen ,  Desmond Yan ,  Blair Mackiewich

发明人： Minghua Chen ,  Desmond Yan ,  Blair Mackiewich

IPC分类号： H04L12/26

CPC分类号： H04L47/22 ,  H04L47/12 ,  H04L47/19 ,  H04L47/20 ,  H04L47/2408 ,  H04L47/2416 ,  H04L47/2441 ,  H04L47/29 ,  H04L47/30 ,  H04L47/39

摘要： Apparatus and techniques for scheduling packets to reduce congestion in a packet data network are described. In one embodiment of the invention, packets are received through each of a plurality of channels and are stored in a respective queue for each channel. The status of each queue is monitored as empty or not empty. When an empty queue becomes not empty, then bandwidth allocation value is added for the respective channel to a port load value. When a not empty queue becomes empty, then the bandwidth allocation for the respective channel is subtracted from the port load value. The traffic rate through the port is then scaled based upon the port load value.

摘要翻译： 描述了用于调度分组以减少分组数据网络中的拥塞的装置和技术。 在本发明的一个实施例中，分组通过多个信道中的每一个被接收，并且被存储在每个信道的相应队列中。 每个队列的状态被监视为空或不为空。 当空队列不为空时，将相应通道的带宽分配值添加到端口负载值。 当不为空的队列变空时，从端口负载值中减去相应通道的带宽分配。 然后通过端口的流量速率根据端口负载值进行缩放。

公开(公告)号：US09058166B2

公开(公告)日：2015-06-16

申请号：US13412521

申请日：2012-03-05

申请人： Desmond Yan ,  Thomas Ng

发明人： Desmond Yan ,  Thomas Ng

IPC分类号： G06F9/46 ,  G06F1/32 ,  G06F9/50

CPC分类号： G06F1/3206 ,  G06F1/3287 ,  G06F9/5094 ,  Y02D10/171 ,  Y02D10/22 ,  Y02D50/20

摘要： In one aspect, the present invention reduces average power consumption in a distributed processing system by concentrating an overall processing load to the minimum number of processing units required to maintain a defined level of processing redundancy. When the required number of active processing units is fewer than all available processing units, the inactive processing units may be held in a reduced-power condition. The present invention thereby maintains the defined level of processing redundancy for reallocating jobs responsive to the failure of one of the active processing units, while reducing power consumption and simplifying jobs allocation and re-allocation when expanding or shrinking the active set of processing units responsive to changing processing load. As a non-limiting example, the distributed processing system is implemented within a telecommunications network router or other apparatus having a configured set of processing cards, such as control-plane processing cards.

摘要翻译： 一方面，本发明通过将整体处理负荷集中在维持规定的处理冗余水平所需的最小数量的处理单元来降低分布式处理系统中的平均功耗。 当所需数量的活动处理单元少于所有可用处理单元时，非活动处理单元可以保持在降低功率状态。 因此，本发明可以响应于一个活动处理单元的故障而维持用于重新分配作业的定义的处理冗余级别，同时在扩展或缩小有效的一组处理单元时降低功耗并简化作业分配和重新分配 改变处理负荷。 作为非限制性示例，分布式处理系统在具有配置的处理卡集合（诸如控制平面处理卡）的电信网路由器或其他设备内实现。

公开(公告)号：US08542586B2

公开(公告)日：2013-09-24

申请号：US13237745

申请日：2011-09-20

申请人： Minghua Chen ,  Desmond Yan

发明人： Minghua Chen ,  Desmond Yan

IPC分类号： H04L12/26 ,  H04L12/28 ,  H04L12/56 ,  G06F11/00 ,  H04Q7/20

CPC分类号： H04L47/125 ,  H04L47/10 ,  H04L47/22

摘要： A method and apparatus to shape packet traffic rates of packets in-transit from a plurality of channels to a port, monitor a status of each queue, subtract a CIR from a port load value when an empty queue becomes active, add the CIR to the port load value when an active queue becomes empty, add a difference between an EIR and the CIR for a channel to a port excess value when a queue of the channel becomes excess, subtract the difference for the channel from the port excess value when the excess queue for the channel becomes not excess, calculate an oversubscription ratio based on the port load value and port excess value, calculate a shaping rate for each of the plurality of channels, and shape the traffic for each of the plurality channels using the shaping rate.

摘要翻译： 一种将从多个信道转发到的分组的分组业务速率整形到端口的方法和装置，监视每个队列的状态，当空队列变为活动时从端口负载值中减去CIR，将CIR添加到 端口负载值当活动队列变空时，当通道的队列过多时，EIR和通道的CIR之间的差异会增加到端口超出值，当超出该值时，减去通道与端口超出值的差值 通道的队列变得不够多，基于端口负载值和端口过剩值计算超额预订比率，计算多个信道中的每个信道的整形速率，并且使用整形率对多个信道中的每个信道进行整形。