-
公开(公告)号:US10917340B2
公开(公告)日:2021-02-09
申请号:US16243409
申请日:2019-01-09
发明人: Rakesh Gandhi , Clarence Filsfils , Sagar Soni , Patrick Khordoc
IPC分类号: G06F11/00 , H04L12/721 , H04L12/723 , H04L29/06 , H04L12/26 , H04L12/813 , H04L12/741
摘要: Techniques for in-situ passive performance measurement are described. In one embodiment, a method includes receiving a data packet at a first network element, determining whether measurement information is to be collected for the data packet, providing one or more measurement fields for the data packet based on a determination that measurement information is to be collected for the data packet in which at least one measurement field identifies a measurement type, and forwarding the data packet to a second network element. The method further includes determining, by the second network element, the measurement type for the data packet, and performing one or more actions based on the measurement type.
-
公开(公告)号:US20190280927A1
公开(公告)日:2019-09-12
申请号:US16191676
申请日:2018-11-15
发明人: Clarence Filsfils , Rakesh Gandhi , Zafar Ali
IPC分类号: H04L12/24 , H04L12/26 , H04L12/707 , H04L12/851 , H04L12/715
摘要: Mechanisms are provided for “direct mode” (in-band) performance loss measurement in computer networks where packet loss is measured directly in the data plane using techniques based on ECMP paths. Counters on each of an ingress node and an egress node are configured by a network controller to count traffic for indicators associated with different equal-cost multi-path (ECMP) paths through a network between ingress and egress nodes. Each indicator is toggled on or off during a measurement interval during which traffic is measured by the traffic counters on the ingress and egress nodes for each ECMP path. The traffic counters (measured in bytes/packets) from the ingress and egress nodes are sent via event driven telemetry to the network controller for performance loss measurement determination.
-
公开(公告)号:US20150117222A1
公开(公告)日:2015-04-30
申请号:US14067023
申请日:2013-10-30
发明人: Tarek Saad , Muthurajah Sivabalan , Rakesh Gandhi
IPC分类号: H04L12/729
CPC分类号: H04L45/125 , H04L45/1283 , H04L45/22
摘要: At a first network device, a plurality of paths through a network from a source network device to a destination network device are determined. A vacant bandwidth is calculated for each of the plurality of paths. A primary path is selected from the plurality of paths based on the vacant bandwidth, and a standby path is selected from the plurality of paths based on the vacant bandwidth.
摘要翻译: 在第一网络设备中,确定通过网络从源网络设备到目的地网络设备的多条路径。 为多个路径中的每一个计算空闲带宽。 基于空闲带宽从多个路径选择主路径,并且基于空闲带宽从多个路径中选择备用路径。
-
公开(公告)号:US20230135261A1
公开(公告)日:2023-05-04
申请号:US18147158
申请日:2022-12-28
发明人: Clarence Filsfils , Zafar Ali , Syed Kamran Raza , Ahmed Bashandy , Nagendra Kumar Nainar , Carlos M. Pignataro , Jaganbabu Rajamanickam , Rakesh Gandhi , Bhupendra Yadav , Faisal Iqbal
IPC分类号: H04L45/00 , H04L43/106 , H04L45/02 , H04L41/0246
摘要: In one embodiment, a service chain data packet is instrumented as it is communicated among network nodes in a network providing service-level and/or networking operations visibility. The service chain data packet includes a particular header identifying a service group defining one or more service functions, and is a data packet and not a probe packet. A network node adds networking and/or service-layer operations data to the particular service chain data packet, such as, but not limited to, in the particular header. Such networking operations data includes a performance metric or attribute related to the transport of the particular service chain packet in the network. Such service-layer operations data includes a performance metric or attribute related to the service-level processing of the particular service chain data packet in the network.
-
公开(公告)号:US11184235B2
公开(公告)日:2021-11-23
申请号:US16191676
申请日:2018-11-15
发明人: Clarence Filsfils , Rakesh Gandhi , Zafar Ali
IPC分类号: H04L12/24 , H04L12/26 , H04L12/851 , H04L12/715 , H04L12/707 , H04L29/06
摘要: Mechanisms are provided for “direct mode” (in-band) performance loss measurement in computer networks where packet loss is measured directly in the data plane using techniques based on ECMP paths. Counters on each of an ingress node and an egress node are configured by a network controller to count traffic for indicators associated with different equal-cost multi-path (ECMP) paths through a network between ingress and egress nodes. Each indicator is toggled on or off during a measurement interval during which traffic is measured by the traffic counters on the ingress and egress nodes for each ECMP path. The traffic counters (measured in bytes/packets) from the ingress and egress nodes are sent via event driven telemetry to the network controller for performance loss measurement determination.
-
公开(公告)号:US20210111999A1
公开(公告)日:2021-04-15
申请号:US17130336
申请日:2020-12-22
发明人: Rakesh Gandhi , Clarence Filsfils , Sagar Soni , Patrick Khordoc
IPC分类号: H04L12/721 , H04L12/723 , H04L29/06 , H04L12/26 , H04L12/813 , H04L12/741
摘要: Techniques for in-situ passive performance measurement are described. In one embodiment, a method includes receiving a data packet at a first network element, determining whether measurement information is to be collected for the data packet, providing one or more measurement fields for the data packet based on a determination that measurement information is to be collected for the data packet in which at least one measurement field identifies a measurement type, and forwarding the data packet to a second network element. The method further includes determining, by the second network element, the measurement type for the data packet, and performing one or more actions based on the measurement type.
-
7.
公开(公告)号:US20200076719A1
公开(公告)日:2020-03-05
申请号:US16280962
申请日:2019-02-20
发明人: Clarence Filsfils , Rakesh Gandhi , Tarek Saad , Sagar Soni , Patrick Khordoc
摘要: The present technology is directed to a scalable solution for end-to-end performance delay measurement for Segment Routing Policies on both SR-MPLS and SRv6 data planes. The scalability of the solution stems from the use of distributed PM sessions along SR Policy ECMP paths. This is achieved by dividing the SR policy into smaller sections comprised of SPT trees or sub-paths, each of which is associated with a Root-Node. Downstream SID List TLVs may be used in Probe query messages for signaling SPT information to the Root-Nodes Alternatively, this SPT signaling may be accomplished by using a centralized controller. Root-Nodes are responsible for dynamically creating PM sessions and measuring delay metrics for their associated SPT tree section. The root-nodes then send the delay metrics for their local section to an ingress PE node or to a centralized controller using delay metric TLV field of the response message.
-
公开(公告)号:US10469367B2
公开(公告)日:2019-11-05
申请号:US15841276
申请日:2017-12-13
发明人: Clarence Filsfils , Zafar Ali , Syed Kamran Raza , Ahmed Refaat Bashandy , Nagendra Kumar Nainar , Carlos M. Pignataro , Jaganbabu Rajamanickam , Rakesh Gandhi , Bhupendra Yadav , Faisal Iqbal
IPC分类号: H04L12/721 , H04L12/26 , H04L12/751 , H04L12/733 , H04L29/12 , H04L12/725 , H04L12/749
摘要: In one embodiment, segment routing (SR) network processing of packets is performed which includes operations signaling and processing of packets in manners providing processing and/or memory efficiencies. One embodiment includes acquiring a segment routing particular packet by a particular router in a network. Responsive to the particular router data plane ascertained during fast path processing by a fast path processing unit that the segment routing particular packet is to be Operations, Administration, and Maintenance (OAM) processed by a different processing unit in the particular router, communicating a time stamp of a current time and the segment routing particular packet including a segment routing header that includes OAM signaling from said fast path processing to the different processing unit, with fast path processing being hardware-based packet processing by the fast path processing unit. The segment routing particular packet is OAM processing by the different processing unit.
-
公开(公告)号:US20230031423A1
公开(公告)日:2023-02-02
申请号:US17565823
申请日:2021-12-30
发明人: Rakesh Gandhi , Clarence Filsfils
IPC分类号: H04L45/12 , H04L45/50 , H04L43/0852
摘要: Techniques are described for utilizing entropy labels of a Multiprotocol Label Switching (MPLS) label stack for performing monitoring operations (e.g., telemetry, performance measurement, OAM, etc.) without altering the MPLS label stack and/or packet path (e.g., ECMP path). The techniques may include determining, by a node of a network, to perform a monitoring operation associated with traffic that is to be sent along a path through the network. In some examples, the node may receive a packet that is to be sent along the path and encapsulate the packet with an MPLS header. The MPLS header may include an entropy label, entropy label indicator, or other label that is capable of carrying a flag indicating the monitoring operation to be performed. The flag may be carried in a TTL field or traffic class field of the label such that the MPLS label stack is not altered to trigger the monitoring operation.
-
公开(公告)号:US20230021278A1
公开(公告)日:2023-01-19
申请号:US17741241
申请日:2022-05-10
发明人: Clarence Filsfils , Zafar Ali , Francois Clad , Bertrand Herbert Guy Duvivier , Mikhail Koldychev , Rakesh Gandhi , Sagar Soni
摘要: Disclosed are systems, apparatuses, methods, and computer-readable media to implement circuit-style network with co-routed bidirectional network paths. A method includes receiving a request for a circuit policy between a source node and a destination node, the circuit policy defining a co-routed bidirectional policy between the source node and the destination node; requesting a path compute service to identify a path between the source node and the destination node that satisfies the circuit policy through a first network; receiving a path identifying a first set of network nodes that satisfy the circuit policy; configuring each node in the first set of network nodes within the first network with the circuit policy; and establishing a connection using the path that satisfies the circuit policy between the source node and the destination node.
-
-
-
-
-
-
-
-
-