US 12,068,946 B2
Interpreting network path state transitions and characteristics
Mukund Yelahanka Raghuprasad, San Jose, CA (US); Jean-Philippe Vasseur, Saint Martin d'Uriage (FR); Grégory Mermoud, Venthône (CH); and Petar Stupar, Lausanne (CH)
Assigned to Cisco Technology, Inc., San Jose, CA (US)
Filed by Cisco Technology, Inc., San Jose, CA (US)
Filed on Nov. 8, 2022, as Appl. No. 17/982,957.
Prior Publication US 2024/0154898 A1, May 9, 2024
Int. Cl. H04L 45/12 (2022.01); H04L 43/0817 (2022.01); H04L 45/24 (2022.01)
CPC H04L 45/123 (2013.01) [H04L 43/0817 (2013.01); H04L 45/24 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method, comprising:
forming, by a device, a latent space by applying dimensionality reduction to timeseries snippets of path metrics for a network path via which traffic for an online application is conveyed;
extracting, by the device, a plurality of path states from the latent space;
presenting, by the device, the plurality of path states for display by a user interface;
receiving, at the device and from the user interface, an indication that a particular network state is an early sign of application experience provided by the online application becoming degraded;
associating, by the device and based on the indication, a set of transitions between the plurality of path states with the online application providing degraded application experience;
using, by the device, the set of transitions to make a prediction as to when the online application will provide the degraded application experience; and
causing, by the device and based on the prediction, the traffic for the online application to be rerouted away from the network path.