# US 10,454,828 B2 · Machine learning-derived entropy path graph from in-situ OAM (iOAM) data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,454,828 B2 - Document ID: US10454828B2 - Status: Granted - Kind: B2 - Application number: 15/386,306 - Filed: 2016-12-21 - Published: 2018-06-21 - Granted: 2019-10-22 - Earliest claimed: 2016-12-21 - Assignees: Cisco Technology, Inc. - Applicants: Cisco Technology, Inc. - Inventors: Carlos M. Pignataro; Nagendra Kumar Nainar; Rajiv Asati; K. Tirumaleswar Reddy - Web page: https://patentgrep.com/patent/US10454828B2 ## CPC - H04L 41/5019: Ensuring fulfilment of SLA - H04L 41/12: Discovery or management of network topologies - H04L 43/14: {using software, i.e. software packages (network security related monitoring H04L63/1408)} - H04L 45/24: Multipath - H04L 47/12: Avoiding congestion; Recovering from congestion - H04L 47/829: {Topology based} - H04L 67/10: in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) - H04L 67/12: specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks - H04L 69/22: Parsing or analysis of headers - H04L 43/045: for graphical visualisation of monitoring data - H04L 43/10: Active monitoring, e.g. heartbeat, ping or trace-route - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract In one embodiment, a device in a network receives in-situ operations administration and management (iOAM) data regarding a plurality of traffic flows in the network. The iOAM data comprises entropy values for the plurality of traffic flows. The device receives network topology information indicative of network paths available in the network. The device generates a machine learning-based entropy topology model for the network based on the received iOAM data and the received network topology information. The entropy topology model maps path selection predictions for the network paths with entropy values. The device uses the entropy topology model to cause a particular traffic flow to use a particular network path. ## Claims ### Claim 1 A method comprising: receiving, at a device in a network, in-situ operations administration and management (iOAM) data packet regarding a plurality of traffic flows in the network, wherein the iOAM data packet comprises entropy values for the plurality of traffic flows; receiving, at the device, network topology information indicative of network paths available in the network; generating, by the device, a machine learning-based entropy topology model for the network based on the received iOAM data packet and the received network topology information, wherein the entropy topology model maps path selection predictions for the network paths with the entropy values; and using, by the device, the entropy topology model to cause a particular traffic flow to use a particular network path, wherein the network uses Equal-Cost Multi-Path (ECMP) routing to route the plurality of traffic flows via the network based in part on the entropy values. ### Claim 2 (depends on 1) The method as in claim 1, wherein using the entropy topology model to cause the particular traffic flow to use a particular network path comprises: causing, by the device, a selected entropy value to be inserted into a header of the particular traffic flow when ingressing the network based on the path selection prediction for the particular network path that is mapped with the selected entropy value. ### Claim 3 (depends on 1) The method as in claim 1, further comprising: receiving, at the device, probing data from out-of-band probes sent via the network; and using, by the device, the probing data to validate the entropy topology model. ### Claim 4 (depends on 3) The method as in claim 3, further comprising: updating, by the device, the entropy topology model based on the probing data. ### Claim 5 (depends on 1) The method as in claim 1, wherein the path selection predictions are associated with time periods. ### Claim 6 (depends on 1) The method as in claim 1, wherein the particular network path is selected in part on a service level agreement (SLA) associated with the particular traffic flow. ### Claim 7 (depends on 1) The method as in claim 1, wherein the particular network path is selected to avoid congestion in the network. ### Claim 8 (depends on 1) The method as in claim 1, wherein the iOAM data packet is gathered using headers of the plurality of traffic flows, and wherein the traffic flows convey user traffic in the network. ### Claim 9 (depends on 1) The method as in claim 1, wherein the network topology information comprises Interior Gateway Protocol (IGP) database information or Border Gateway Protocol-Link State (BGP-LS) information. ### Claim 10 (depends on 1) The method as in claim 1, wherein the entropy values are random labels included in a header field of the iOAM data packet. ### Claim 11 An apparatus, comprising: one or more network interfaces to communicate with a segment routed network; a processor coupled to the one or more network interfaces and configured to execute one or more processes; and a memory configured to store a process executable by the processor, the process when executed operable to: receive in-situ operations administration and management (iOAM) data packet regarding a plurality of traffic flows in the network, wherein the iOAM data packet comprises entropy values for the plurality of traffic flows; receive network topology information indicative of network paths available in the network; generate a machine learning-based entropy topology model for the network based on the received iOAM data packet and the received network topology information, wherein the entropy topology model maps path selection predictions for the network paths with the entropy values; and use the entropy topology model to cause a particular traffic flow to use a particular network path, wherein the network uses Equal-Cost Multi-Path (ECMP) routing to route the plurality of traffic flows via the network based in part on the entropy values. ### Claim 12 (depends on 11) The apparatus as in claim 11, wherein the apparatus uses the entropy topology model to cause the particular traffic flow to use a particular network path by: causing a selected entropy value to be inserted into a header of the particular traffic flow when ingressing the network based on the path selection prediction for the particular network path that is mapped with the selected entropy value. ### Claim 13 (depends on 11) The apparatus as in claim 11, wherein the process when executed is further operable to: receive probing data from out-of-band probes sent via the network; and use the probing data to validate the entropy topology model. ### Claim 14 (depends on 11) The apparatus as in claim 11, wherein the path selection predictions are associated with time periods. ### Claim 15 (depends on 11) The apparatus as in claim 11, wherein the particular network path is selected in part on a service level agreement (SLA) associated with the particular traffic flow. ### Claim 16 (depends on 11) The apparatus as in claim 11, wherein the particular network path is selected to avoid congestion in the network. ### Claim 17 (depends on 11) The apparatus as in claim 11, wherein the iOAM data packet is gathered using headers of the plurality of traffic flows, and wherein the traffic flows convey user traffic in the network. ### Claim 18 (depends on 11) The apparatus as in claim 11, wherein the network topology information comprises Interior Gateway Protocol (IGP) database information or Border Gateway Protocol-Link State (BGP-LS) information. ### Claim 19 (depends on 11) The apparatus as in claim 11, wherein the entropy values are random labels included in a header field of the iOAM data packet. ### Claim 20 A tangible, non-transitory, computer-readable media having software encoded thereon, the software when executed by a device in a network operable to perform a process comprising: receiving, at the device, in-situ operations administration and management (iOAM) data packet regarding a plurality of traffic flows in the network, wherein the iOAM data packet comprises entropy values for the plurality of traffic flows; receiving, at the device, network topology information indicative of network paths available in the network; generating, by the device, a machine learning-based entropy topology model for the network based on the received iOAM data packet and the received network topology information, wherein the entropy topology model maps path selection predictions for the network paths with the entropy values; and using, by the device, the entropy topology model to cause a particular traffic flow to use a particular network path, wherein the network uses Equal-Cost Multi-Path (ECMP) routing to route the plurality of traffic flows via the network based in part on the entropy values. ### Claim 21 (depends on 20) The tangible, non-transitory, computer-readable media as in claim 20, wherein the entropy values are random labels included in a header field of the iOAM data packet.