US 11,916,777 B2
Learning SLA violation probability from intelligent fine grained probing
Jean-Philippe Vasseur, Saint Martin d'Uriage (FR); Grégory Mermoud, Venthône (CH); Vinay Kumar Kolar, San Jose, CA (US); David Tedaldi, Zurich (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 Jul. 6, 2021, as Appl. No. 17/368,165.
Prior Publication US 2023/0008106 A1, Jan. 12, 2023
Int. Cl. H04L 45/00 (2022.01); H04L 43/10 (2022.01); H04L 43/0829 (2022.01); H04L 43/087 (2022.01); H04L 43/0888 (2022.01); H04L 43/12 (2022.01)
CPC H04L 45/22 (2013.01) [H04L 43/087 (2013.01); H04L 43/0829 (2013.01); H04L 43/0888 (2013.01); H04L 43/10 (2013.01); H04L 43/12 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining, by a device, a first set of measurements of a path metric for a path in a network that are measured using periodic probing of the path;
obtaining, by the device, a second set of measurements of the path metric for the path that are measured using fine-grained probing of the path at a higher frequency than that of the periodic probing by providing a trigger pattern comprising a set of feature vectors to a router associated with the path, wherein the router initiates the fine-grained probing of the path when measurements in the first set of measurements match the trigger pattern;
generating, by the device, a predictive model that predicts values of the path metric, based on the first set of measurements and on the second set of measurements; and
causing, by the device and based on a value of the path metric predicted by the predictive model, traffic to be rerouted from the path to another path in the network,
wherein the second set of measurements is obtained based on a control strategy that optimizes an accuracy of the predictive model while minimizing an amount of the fine-grained probing performed along the path.