-
1.发明申请公开(公告)号：US20090161684A1
公开(公告)日：2009-06-25
申请号：US11963293
申请日：2007-12-21
申请人： Sreenivas Voruganti , Atul Mahamuni
发明人： Sreenivas Voruganti , Atul Mahamuni
IPC分类号： H04L12/56
CPC分类号： H04L49/9078 , H04L49/90
摘要： Methods and systems consistent with the present invention provide dynamic buffer allocation to a plurality of queues of differing priority levels. Each queue is allocated fixed minimum number of buffers that will not be de-allocated during buffer reassignment. The rest of the buffers are intelligently and dynamically assigned to each queue depending on their current need. The system then monitors and learns the incoming traffic pattern and resulting drops in each queue due to traffic bursts. Based on this information, the system readjusts allocation of buffers to each traffic class. If a higher priority queue does not need the buffers, it gradually relinquishes them. These buffers are then assigned to other queues based on the input traffic pattern and resultant drops. These buffers are aggressively reclaimed and reassigned to higher priority queues when needed. In this way, methods and systems consistent with the present invention dynamically balance requirements of the higher priority queues versus optimal allocation.
摘要翻译： 与本发明一致的方法和系统为不同优先级的多个队列提供动态缓冲器分配。 每个队列被分配固定的最小数量的缓冲区，在缓冲区重新分配期间不会被分配。 其余的缓冲区根据当前的需要智能动态地分配给每个队列。 然后系统监视和学习由于流量突发而导致的每个队列中的进入流量模式和结果丢失。 基于此信息，系统重新调整每个流量级别的缓冲区分配。 如果优先级较高的队列不需要缓冲区，则逐渐放弃它们。 然后，这些缓冲区将根据输入流量模式和结果丢失分配给其他队列。 这些缓冲区在需要时被积极回收并重新分配给较高优先级的队列。 以这种方式，与本发明一致的方法和系统动态地平衡较高优先级队列与最佳分配的要求。
-
2.发明授权公开(公告)号：US08160085B2
公开(公告)日：2012-04-17
申请号：US11963293
申请日：2007-12-21
申请人： Sreenivas Voruganti , Atul Mahamuni
发明人： Sreenivas Voruganti , Atul Mahamuni
IPC分类号： H04L12/54
CPC分类号： H04L49/9078 , H04L49/90
摘要： Methods and systems consistent with the present invention provide dynamic buffer allocation to a plurality of queues of differing priority levels. Each queue is allocated fixed minimum number of buffers that will not be de-allocated during buffer reassignment. The rest of the buffers are intelligently and dynamically assigned to each queue depending on their current need. The system then monitors and learns the incoming traffic pattern and resulting drops in each queue due to traffic bursts. Based on this information, the system readjusts allocation of buffers to each traffic class. If a higher priority queue does not need the buffers, it gradually relinquishes them. These buffers are then assigned to other queues based on the input traffic pattern and resultant drops. These buffers are aggressively reclaimed and reassigned to higher priority queues when needed. In this way, methods and systems consistent with the present invention dynamically balance requirements of the higher priority queues versus optimal allocation.
摘要翻译： 与本发明一致的方法和系统为不同优先级的多个队列提供动态缓冲器分配。 每个队列被分配固定的最小数量的缓冲区，在缓冲区重新分配期间不会被分配。 其余的缓冲区根据当前的需要智能动态地分配给每个队列。 然后系统监视和学习由于流量突发而导致的每个队列中的进入流量模式和结果丢失。 基于此信息，系统重新调整每个流量级别的缓冲区分配。 如果优先级较高的队列不需要缓冲区，则逐渐放弃它们。 然后，这些缓冲区将根据输入流量模式和结果丢失分配给其他队列。 这些缓冲区在需要时被积极回收并重新分配给较高优先级的队列。 以这种方式，与本发明一致的方法和系统动态地平衡较高优先级队列与最佳分配的要求。
-
3.发明申请公开(公告)号：US20120275464A1
公开(公告)日：2012-11-01
申请号：US13447660
申请日：2012-04-16
申请人： Sreenivas Voruganti , Atul Mahamuni
发明人： Sreenivas Voruganti , Atul Mahamuni
IPC分类号： H04L12/54
CPC分类号： H04L49/9078 , H04L49/90
摘要： Methods and systems consistent with the present invention provide dynamic buffer allocation to a plurality of queues of differing priority levels. Each queue is allocated fixed minimum number of buffers that will not be de-allocated during buffer reassignment. The rest of the buffers are intelligently and dynamically assigned to each queue depending on their current need. The system then monitors and learns the incoming traffic pattern and resulting drops in each queue due to traffic bursts. Based on this information, the system readjusts allocation of buffers to each traffic class. If a higher priority queue does not need the buffers, it gradually relinquishes them. These buffers are then assigned to other queues based on the input traffic pattern and resultant drops. These buffers are aggressively reclaimed and reassigned to higher priority queues when needed.
摘要翻译： 与本发明一致的方法和系统为不同优先级的多个队列提供动态缓冲器分配。 每个队列被分配固定的最小数量的缓冲区，在缓冲区重新分配期间不会被分配。 其余的缓冲区根据当前的需要智能动态地分配给每个队列。 然后系统监视和学习由于流量突发而导致的每个队列中的进入流量模式和结果丢失。 基于此信息，系统重新调整每个流量级别的缓冲区分配。 如果优先级较高的队列不需要缓冲区，则逐渐放弃它们。 然后，这些缓冲区将根据输入流量模式和结果丢失分配给其他队列。 这些缓冲区在需要时被积极回收并重新分配给较高优先级的队列。
-
4.发明授权公开(公告)号：US08848724B2
公开(公告)日：2014-09-30
申请号：US13447660
申请日：2012-04-16
申请人： Sreenivas Voruganti , Atul Mahamuni
发明人： Sreenivas Voruganti , Atul Mahamuni
IPC分类号： H04L12/28 , H04L12/861
CPC分类号： H04L49/9078 , H04L49/90
摘要： Methods and systems consistent with the present invention provide dynamic buffer allocation to a plurality of queues of differing priority levels. Each queue is allocated fixed minimum number of buffers that will not be de-allocated during buffer reassignment. The rest of the buffers are intelligently and dynamically assigned to each queue depending on their current need. The system then monitors and learns the incoming traffic pattern and resulting drops in each queue due to traffic bursts. Based on this information, the system readjusts allocation of buffers to each traffic class. If a higher priority queue does not need the buffers, it gradually relinquishes them. These buffers are then assigned to other queues based on the input traffic pattern and resultant drops. These buffers are aggressively reclaimed and reassigned to higher priority queues when needed.
摘要翻译： 与本发明一致的方法和系统为不同优先级的多个队列提供动态缓冲器分配。 每个队列被分配固定的最小数量的缓冲区，在缓冲区重新分配期间不会被分配。 其余的缓冲区根据当前的需要智能动态地分配给每个队列。 然后系统监视和学习由于流量突发而导致的每个队列中的进入流量模式和结果丢失。 基于此信息，系统重新调整每个流量级别的缓冲区分配。 如果优先级较高的队列不需要缓冲区，则逐渐放弃它们。 然后，这些缓冲区将根据输入流量模式和结果丢失分配给其他队列。 这些缓冲区在需要时被积极回收并重新分配给较高优先级的队列。
-
5.发明申请QUALITY OF SERVICE (QOS) CONFIGURATION FOR NETWORK DEVICES WITH MULTIPLE QUEUES 审中-公开服务质量（QOS）具有多个队列的网络设备配置公开(公告)号：US20120243412A1
公开(公告)日：2012-09-27
申请号：US13485363
申请日：2012-05-31
摘要： A network device implements automatic configuration of Quality of Service (QoS) parameters in response to operator specification of a relatively few and easily understandable “high level” parameters such as, for example, latency requirements or an acceptable rate of packet drops. In one implementation, a network device may receive user preference information that relates to a Quality of Service (QoS) for network traffic passing through the network device and may measure traffic patterns through the network device. The device further generates a configuration template based on the measured traffic patterns and on the user preference information transmit the data in an order of transmission that is prioritized according to a bandwidth allocation policy defined by the configuration template.
摘要翻译： 响应于操作员指定相对较少且易于理解的“高级”参数（例如等待时间要求或可接受的分组丢弃率），网络设备实现服务质量（QoS）参数的自动配置。 在一个实现中，网络设备可以接收与通过网络设备的网络业务的服务质量（QoS）相关的用户偏好信息，并且可以通过网络设备测量流量模式。 该设备还基于所测量的流量模式进一步生成配置模板，并且用户偏好信息根据由配置模板定义的带宽分配策略以传输顺序发送数据。
-
6.发明授权公开(公告)号：US08144584B1
公开(公告)日：2012-03-27
申请号：US12505992
申请日：2009-07-20
IPC分类号： G01R31/08
CPC分类号： H04L47/6225 , H04L47/522 , H04L47/6215
摘要： A method includes receiving network information for calculating weighted round-robin (WRR) weights, calculating WRR weights associated with queues based on the network information, and determining whether a highest common factor (HCF) exists in relation to the calculated WRR weights. The method further includes reducing the calculated WRR weights in accordance with the HCF, when it is determined that the HCF exists, and performing a WRR scheduling of packets, stored in the queues, based on the reduced WRR weights.
摘要翻译： 一种方法，包括接收网络信息，用于计算加权轮询（WRR）权重，基于网络信息计算与队列相关联的WRR权重，以及确定是否存在与所计算的WRR权重有关的最高公因子（HCF）。 该方法还包括当确定HCF存在时，根据HCF减少所计算的WRR权重，并且基于减少的WRR权重对存储在队列中的分组执行WRR调度。
-
7.发明授权Dynamic prioritized fair share scheduling scheme in over-subscribed port scenario 有权超订购端口情景下动态优先公平共享调度方案公开(公告)号：US09054996B2
公开(公告)日：2015-06-09
申请号：US12647048
申请日：2009-12-24
IPC分类号： H04L12/931 , H04L12/801 , H04L12/813 , H04L12/835
CPC分类号： H04L49/505 , H04L47/12 , H04L47/20 , H04L47/29 , H04L47/30 , H04L49/506
摘要： A network device receives initial policer limits for a plurality of over-subscribing ingress ports, where the initial policer limits are based on existing bandwidth limits for an over-subscribed egress port associated with the over-subscribing ingress ports. The network device receives a high threshold watermark and a low threshold watermark for bandwidth usage of the over-subscribed egress port, and identifies a queue, associated with the over-subscribed egress port, with values outside the high threshold watermark or the low threshold watermark. The network device reduces the initial policer limits for the plurality of over-subscribing ingress ports when the queue has values above the high threshold watermark, and increases the initial policer limits for the plurality of over-subscribing ingress ports when the queue has values below the low threshold watermark.
摘要翻译： 网络设备接收多个过度订阅入口端口的初始策略器限制，其中初始策略器限制基于与超订阅入口端口相关联的超订购出口端口的现有带宽限制。 网络设备接收高阈值水印和低阈值水印以用于过度订阅的出口端口的带宽使用，并且用高阈值水印或低阈值水印之外的值识别与超订阅出口端口相关联的队列 。 当队列具有高于高阈值水印的值时，网络设备降低多个超订阅入口端口的初始策略器限制，并且当队列具有低于阈值水印值时，增加多个超订阅入口端口的初始策略器限制 低阈值水印。
-
8.发明授权Quality of service (QoS) configuration for network devices with multiple queues 有权具有多个队列的网络设备的服务质量（QoS）配置公开(公告)号：US08879393B2
公开(公告)日：2014-11-04
申请号：US13485363
申请日：2012-05-31
IPC分类号： G01R31/08 , H04L12/54 , H04L12/823 , H04L12/835
摘要： A network device implements automatic configuration of Quality of Service (QoS) parameters in response to operator specification of a relatively few and easily understandable “high level” parameters such as, for example, latency requirements or an acceptable rate of packet drops. In one implementation, a network device may receive user preference information that relates to a Quality of Service (QoS) for network traffic passing through the network device and may measure traffic patterns through the network device. The device further generates a configuration template based on the measured traffic patterns and on the user preference information transmit the data in an order of transmission that is prioritized according to a bandwidth allocation policy defined by the configuration template.
摘要翻译： 响应于操作员指定相对较少且易于理解的“高级”参数（例如等待时间要求或可接受的分组丢弃率），网络设备实现服务质量（QoS）参数的自动配置。 在一个实现中，网络设备可以接收与通过网络设备的网络业务的服务质量（QoS）相关的用户偏好信息，并且可以通过网络设备测量流量模式。 该设备还基于所测量的流量模式进一步生成配置模板，并且用户偏好信息根据由配置模板定义的带宽分配策略以传输顺序发送数据。
-
9.发明授权Quality of service (QoS) configuration for network devices with multiple queues 有权具有多个队列的网络设备的服务质量（QoS）配置公开(公告)号：US08213453B1
公开(公告)日：2012-07-03
申请号：US12537525
申请日：2009-08-07
摘要： A network device implements automatic configuration of Quality of Service (QoS) parameters in response to operator specification of a relatively few and easily understandable “high level” parameters such as, for example, latency requirements or an acceptable rate of packet drops. In one implementation, a network device may receive user preference information that relates to a Quality of Service (QoS) for network traffic passing through the network device and may measure traffic patterns through the network device. The device further generates a configuration template based on the measured traffic patterns and on the user preference information transmit the data in an order of transmission that is prioritized according to a bandwidth allocation policy defined by the configuration template.
摘要翻译： 响应于操作员指定相对较少且易于理解的“高级”参数（例如等待时间要求或可接受的分组丢弃率），网络设备实现服务质量（QoS）参数的自动配置。 在一个实现中，网络设备可以接收与通过网络设备的网络业务的服务质量（QoS）相关的用户偏好信息，并且可以通过网络设备测量流量模式。 该设备还基于所测量的流量模式进一步生成配置模板，并且用户偏好信息根据由配置模板定义的带宽分配策略以传输顺序发送数据。
-
10.发明授权公开(公告)号：US08351332B2
公开(公告)日：2013-01-08
申请号：US13371165
申请日：2012-02-10
IPC分类号： G01R31/08
CPC分类号： H04L47/6225 , H04L47/522 , H04L47/6215
摘要： A method includes receiving network information for calculating weighted round-robin (WRR) weights, calculating WRR weights associated with queues based on the network information, and determining whether a highest common factor (HCF) exists in relation to the calculated WRR weights. The method further includes reducing the calculated WRR weights in accordance with the HCF, when it is determined that the HCF exists, and performing a WRR scheduling of packets, stored in the queues, based on the reduced WRR weights.
摘要翻译： 一种方法，包括接收网络信息，用于计算加权轮询（WRR）权重，基于网络信息计算与队列相关联的WRR权重，以及确定是否存在与所计算的WRR权重有关的最高公因子（HCF）。 该方法还包括当确定HCF存在时，根据HCF减少所计算的WRR权重，并且基于减少的WRR权重对存储在队列中的分组执行WRR调度。
-
-
-
-
-
-
-
-
-
搜索结果
12 条记录 (耗时 0.016 秒)
