公开(公告)号：US10250504B1

公开(公告)日：2019-04-02

申请号：US15724092

申请日：2017-10-03

Applicant: Cisco Technology, Inc.

Inventor： Venkata Naga Ramesh Ponnapalli ,  Ayan Banerjee ,  Raghava K Sivaramu ,  Nataraj Bacthu

IPC: H04L12/745 ,  H04L12/743 ,  H04L12/721

Abstract: In one embodiment, longest prefix matching (LPM) operations are performed on a value in multiple interspersed prefix length search spaces to determine an overall longest prefix matching result in a packet switching system. A first LPM lookup unit performs a first LPM operation on the particular lookup value in a first search space finding a first longest matching prefix, and a second LPM lookup unit performs a second LPM operation on the particular lookup value in a second search space finding a second longest matching prefix. The longer of the first and second longest matching prefixes determines the overall LPM. In one embodiment, the first search space and the second search space include non-default route prefixes with interspersed prefix lengths matching a same value, such as, but not limited to the particular lookup value (e.g., a destination address of a packet).

公开(公告)号：US20190097839A1

公开(公告)日：2019-03-28

申请号：US15714761

申请日：2017-09-25

Applicant: Cisco Technology, Inc.

Inventor： Swami Narayanan ,  Ambrish Mehta ,  Venkatesh Srinivasan ,  Raghava Sivaramu ,  Ayan Banerjee

IPC: H04L12/46 ,  H04L29/06 ,  H04L12/04

Abstract: Aspects of the disclosed technology address limitations relating to packet replication for multi-destination traffic, by providing methods for performing hardware-based replication in network infrastructure devices, such as switches. In some aspects, application specific integrated circuits (ASICs) resident in physical devices can be used to perform packet replication. Depending on implementation, a hardware-based replication process can include steps for receiving a first packet that includes a first outer header containing first address information, receiving a second packet including a second outer header containing a hardware replication flag, forwarding the first packet to all virtual tunnel endpoints (VTEPs) connected with the TOR switch, and performing hardware replication for the second packet based on the hardware replication flag to generate one or more unicast packets. Systems and machine readable media are also provided.

公开(公告)号：US10079761B2

公开(公告)日：2018-09-18

申请号：US14477753

申请日：2014-09-04

Applicant: Cisco Technology, Inc.

Inventor： Ayan Banerjee ,  Ramana Mellacheruvu ,  Abhishek Saxena ,  Vishal Jain ,  Luca Cafiero

IPC: H04L12/741 ,  H04L12/58 ,  H04W28/12 ,  H04W72/00 ,  H04L12/721 ,  H04L12/801 ,  H04L12/761 ,  H04L12/745

CPC classification number: H04L45/745 ,  H04L45/16 ,  H04L45/72 ,  H04L45/748 ,  H04L47/33 ,  H04L51/10 ,  H04W28/12 ,  H04W72/005

Abstract: Systems, methods, and non-transitory computer-readable storage media for performing hierarchical routing are disclosed. The method includes identifying routes in a computer network and arranging those routes in two separate routing tables. The first routing table is stored on a first module and the second routing table is stored on a second module.

公开(公告)号：US09825857B2

公开(公告)日：2017-11-21

申请号：US14449007

申请日：2014-07-31

Applicant: Cisco Technology, Inc.

Inventor： Ayan Banerjee ,  Sameer Merchant ,  Ramana Mellacheruvu ,  Ramesh V. N. Ponnapalli ,  Srinivasan Ramabadran

IPC: H04L12/741 ,  H04L12/58 ,  H04W28/12 ,  H04W72/00 ,  H04L12/721 ,  H04L12/801 ,  H04L12/761 ,  H04L12/745

CPC classification number: H04L45/745 ,  H04L45/16 ,  H04L45/72 ,  H04L45/748 ,  H04L47/33 ,  H04L51/10 ,  H04W28/12 ,  H04W72/005

Abstract: Various embodiments are disclosed for increasing Layer-3 LPM (longest prefix match) routing database in a network platform. In some embodiments, chipsets in fabric modules (FMs) can be partitioned into multiple banks. Network traffic can be directed towards a corresponding bank in the FMs by using a LPM table on a line card (LC). Entries in the LPM table on the LC can be programmed either statically or dynamically based upon LPM routes that are dynamically learned.

公开(公告)号：US09306856B2

公开(公告)日：2016-04-05

申请号：US13842195

申请日：2013-03-15

Applicant: Cisco Technology, Inc.

Inventor： Ramkumar Parameswaran ,  Ayan Banerjee ,  Francois Edouard Tallet ,  Varun Chandulal Shah

IPC: H04L12/801 ,  H04L12/803 ,  H04L12/753 ,  H04L12/18 ,  H04L12/721

CPC classification number: H04L47/125 ,  H04L12/185 ,  H04L12/1886 ,  H04L45/123 ,  H04L45/48

Abstract: Embodiments provide a method, network device, and computer program product for optimizing traffic in a link-state network distributed across a plurality of sites. The method, network device, and computer program product include receiving a multi-destination message at a first node within the link-state network. Additionally, the method, network device, and computer program product include identifying a plurality of multi-destination trees within the network, each tree having a respective root node. The method, network device, and computer program product determine a plurality of cost values corresponding to each of the plurality of multi-destination trees, based on one or more links along a shortest path from the first node to the respective root node of each tree. Based on the determined cost values, the method, network device, and computer program product select one or more of the multi-destination trees and transmit the multi-destination message using the selected tree.

Abstract translation: 实施例提供了一种用于优化分布在多个站点上的链路状态网络中的业务的方法，网络设备和计算机程序产品。 该方法，网络设备和计算机程序产品包括在链路状态网络内的第一节点处接收多目的地消息。 此外，方法，网络设备和计算机程序产品包括识别网络内的多个多目的地树，每个树具有相应的根节点。 所述方法，网络设备和计算机程序产品基于沿着从每个树的第一节点到相应根节点的最短路径的一个或多个链路来确定与所述多个目的地树中的每一个对应的多个成本值 。 基于确定的成本值，方法，网络设备和计算机程序产品选择多目的地树中的一个或多个并且使用所选择的树发送多目的地消息。

公开(公告)号：US20160065449A1

公开(公告)日：2016-03-03

申请号：US14472573

申请日：2014-08-29

Applicant: Cisco Technology, Inc.

Inventor： Ayaskant Pani ,  Ayan Banerjee

IPC: H04L12/707 ,  H04L12/729

CPC classification number: H04L45/24 ,  H04L45/125

Abstract: A plurality of equal cost paths through a network from a source node to a destination node are determined. A maximum bandwidth capacity for each link of each of the plurality of equal cost paths is determined, and a smallest capacity link for each of the plurality of equal cost paths is determined from the maximum capacity bandwidths for each link. An aggregated maximum bandwidth from the source node to the destination node is determined by aggregating the smallest capacity links for each of the plurality of equal cost paths. Traffic is sent from the source node along each of the plurality of equal cost paths according to a value of a capacity for the smallest capacity link for each of the plurality of equal cost paths, wherein a total of the sent traffic does not exceed the aggregated maximum bandwidth.

Abstract translation: 确定通过网络从源节点到目的地节点的多条相同的成本路径。 确定多个相同成本路径中的每一个的每个链路的最大带宽容量，并且根据每个链路的最大容量带宽确定用于多个等价路径中的每一个的最小容量链路。 从源节点到目的地节点的聚合最大带宽通过聚合多个相等成本路径中的每一个的最小容量链路来确定。 根据多个等价路径中的每一个的最小容量链路的容量的值，从多个等价路径中的每一个向源节点发送流量，其中所发送的流量的总和不超过聚合的 最大带宽。

公开(公告)号：US20150124806A1

公开(公告)日：2015-05-07

申请号：US14477753

申请日：2014-09-04

Applicant: Cisco Technology, Inc.

Inventor： Ayan Banerjee ,  Ramana Mellacheruvu ,  Abhishek Saxena ,  Vishal Jain ,  Luca Cafiero

IPC: H04L12/741 ,  H04L12/801 ,  H04L12/721

CPC classification number: H04L45/745 ,  H04L45/16 ,  H04L45/72 ,  H04L45/748 ,  H04L47/33 ,  H04L51/10 ,  H04W28/12 ,  H04W72/005

Abstract: Systems, methods, and non-transitory computer-readable storage media for performing hierarchical routing are disclosed. The method includes identifying routes in a computer network and arranging those routes in two separate routing tables. The first routing table is stored on a first module and the second routing table is stored on a second module.

Abstract translation: 公开了用于执行分层路由的系统，方法和非暂时的计算机可读存储介质。 该方法包括识别计算机网络中的路由并将这些路由布置在两个单独的路由表中。 第一路由表存储在第一模块上，第二路由表存储在第二模块上。

公开(公告)号：US20150124684A1

公开(公告)日：2015-05-07

申请号：US14477244

申请日：2014-09-04

Applicant: Cisco Technology, Inc.

Inventor： Ayan Banerjee ,  Srinivasan Ramabadran ,  Mehak Mahajan ,  Raghava Sivaramu ,  Nataraj Bacthu ,  Raja Rao Tadimeti ,  Madhava Rao Cheethirala ,  Ramana Mellacheruvu

IPC: H04W72/00 ,  H04L12/58 ,  H04W28/12

CPC classification number: H04L45/745 ,  H04L45/16 ,  H04L45/72 ,  H04L45/748 ,  H04L47/33 ,  H04L51/10 ,  H04W28/12 ,  H04W72/005

Abstract: A plurality of line cards with each line card having a respective network forwarding engine and a respective outgoing interface (OIF) list and at least one fabric module communicatively coupled with each line card with each fabric module can have a respective network forwarding engine. The local OIF list can be asymmetrically programmed. The network forwarding engine of a line card can be configured to receive a multicast packet, compare a multicast address associate with the received multicast packet with entries in the local OIF list of the line card and forward the received multicast packet to at least one interface associated with the multicast address in response to the comparison resulting in a match.

Abstract translation: 具有每个线路卡的多个线路卡具有相应的网络转发引擎和相应的输出接口（OIF）列表以及与每个线路卡与每个结构模块通信地耦合的至少一个结构模块可以具有相应的网络转发引擎。 本地OIF列表可以非对称编程。 线路卡的网络转发引擎可以被配置为接收多播分组，将与所接收的多播分组关联的多播地址与线路卡的本地OIF列表中的条目进行比较，并将接收到的多播分组转发到至少一个相关联的接口 多播地址响应于比较导致匹配。

公开(公告)号：US20150124633A1

公开(公告)日：2015-05-07

申请号：US14449007

申请日：2014-07-31

Applicant: Cisco Technology, Inc.

Inventor： Ayan Banerjee ,  Sameer Merchant ,  Ramana Mellacheruvu ,  Ramesh V.N. Ponnapalli ,  Srinivasan Ramabadran

IPC: H04L12/741 ,  H04L12/743 ,  H04L12/721

CPC classification number: H04L45/745 ,  H04L45/16 ,  H04L45/72 ,  H04L45/748 ,  H04L47/33 ,  H04L51/10 ,  H04W28/12 ,  H04W72/005

Abstract: Various embodiments are disclosed for increasing Layer-3 LPM (longest prefix match) routing database in a network platform. In some embodiments, chipsets in fabric modules (FMs) can be partitioned into multiple banks. Network traffic can be directed towards a corresponding bank in the FMs by using a LPM table on a line card (LC). Entries in the LPM table on the LC can be programmed either statically or dynamically based upon LPM routes that are dynamically learned.

Abstract translation: 公开了用于在网络平台中增加第3层LPM（最长前缀匹配）路由数据库的各种实施例。 在一些实施例中，织物模块（FM）中的芯片组可以被划分成多个组。 通过使用线路卡（LC）上的LPM表，网络流量可以指向FM中的相应银行。 可以根据动态学习的LPM路由，静态地或动态地对LC中的LPM表中的条目进行编程。

公开(公告)号：US20140269410A1

公开(公告)日：2014-09-18

申请号：US13826572

申请日：2013-03-14

Applicant: CISCO TECHNOLOGY, INC.

Inventor： Varun Shah ,  Ayan Banerjee ,  Dhananjaya Rao ,  Raghava Sivaramu ,  Abhay Roy

IPC: H04L12/24 ,  H04L12/721

CPC classification number: H04L45/12 ,  H04L45/32 ,  H04L45/48

Abstract: Techniques are provided for generating and updating flooding tree paths in a network. At a particular node device in a network, a first flooding tree is generated by performing a first shortest path first (SPF) operation from a first selected node device in the network to a plurality of other node devices in the network. A second flooding tree is generated by performing a second SPF operation from a second selected node device in the network to the plurality of other node devices in the network. A network topology change event is detected in either the first or second flooding tree, and a packet sequence exchange is initiated between the particular node device and another node device in the network in response to the detected network topology change. The first and second flooding trees are then updated based on information obtained during the packet sequence exchange.

Abstract translation: 提供了用于在网络中生成和更新泛洪树路径的技术。 在网络中的特定节点设备处，通过从网络中的第一选择的节点设备到网络中的多个其他节点设备执行第一最短路径优先（SPF）操作来生成第一洪泛树。 通过从网络中的第二所选节点设备向网络中的多个其他节点设备执行第二SPF操作来生成第二洪泛树。 在第一或第二洪泛树中检测到网络拓扑变化事件，并且响应于检测到的网络拓扑变化，在特定节点设备和网络中的另一个节点设备之间发起分组序列交换。 然后基于在分组序列交换期间获得的信息来更新第一和第二洪泛树。