US 12,231,312 B2
Autonomous system bottleneck detection
Jean-Philippe Vasseur, Saint Martin d'Uriage (FR); Vinay Kumar Kolar, San Jose, CA (US); Grégory Mermoud, Venthône (CH); and Pierre-André Savalle, Rueil-Malmaison (FR)
Assigned to Cisco Technology, Inc., San Jose, CA (US)
Filed by Cisco Technology, Inc., San Jose, CA (US)
Filed on May 24, 2021, as Appl. No. 17/328,205.
Prior Publication US 2022/0376998 A1, Nov. 24, 2022
Int. Cl. H04L 43/045 (2022.01); G06F 16/901 (2019.01); H04L 43/08 (2022.01); H04L 43/12 (2022.01); H04L 69/326 (2022.01)
CPC H04L 43/08 (2013.01) [G06F 16/9024 (2019.01); H04L 43/045 (2013.01); H04L 43/12 (2013.01); H04L 69/326 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining, by a supervisory service for a network, user experience metrics for application sessions of an online application as specified by users of the online application regarding their subjective quality of experiences with the applications sessions of the online application;
mapping, by the supervisory service, the application sessions to paths from an edge router that traverse a plurality of autonomous systems that implement Border Gateway Protocol routing;
identifying, by the supervisory service and based in part on the user experience metrics, a particular autonomous system from the plurality of autonomous systems associated with decreased user experience metrics for the online application;
determining, by the supervisory service, a list of one or more alternative paths from the edge router that avoid the particular autonomous system of the network and any other autonomous systems from the plurality of autonomous systems that are forecasted to cause decreased user experience metrics for the online application; and
causing, by the supervisory service, application traffic for the online application to avoid the particular autonomous system and any other autonomous systems from the plurality of autonomous systems that are forecasted to cause decreased user experience metrics for the online application by providing, to the edge router, the list of one or more alternative paths from the edge router that avoid the particular autonomous system of the network and any other autonomous systems from the plurality of autonomous systems that are forecasted to cause decreased user e metrics for the online application.