US 12,088,426 B2
Automating a software-defined wide area network policy for internet of things end points
Balaji Sundararajan, Fremont, CA (US); Vivek Agarwal, Campbell, CA (US); Anand Oswal, Pleasanton, CA (US); Chethan Channappa, San Jose, CA (US); Subhash Kodnad, Dublin, CA (US); and Jeevan Sharma, Fremont, CA (US)
Assigned to CISCO TECHNOLOGY, INC., San Jose, CA (US)
Filed by Cisco Technology, Inc., San Jose, CA (US)
Filed on Aug. 8, 2022, as Appl. No. 17/882,752.
Application 17/882,752 is a continuation of application No. 16/739,442, filed on Jan. 10, 2020, granted, now 11,411,765.
Prior Publication US 2022/0376982 A1, Nov. 24, 2022
Int. Cl. H04L 12/28 (2006.01); G06F 9/455 (2018.01); G16Y 30/10 (2020.01); H04L 9/40 (2022.01); H04L 12/66 (2006.01); H04L 41/0894 (2022.01); H04L 41/14 (2022.01); H04L 41/50 (2022.01); H04L 47/76 (2022.01); H04L 49/00 (2022.01); H04L 67/12 (2022.01); H04W 92/02 (2009.01)
CPC H04L 12/2856 (2013.01) [G06F 9/45558 (2013.01); G16Y 30/10 (2020.01); H04L 12/2854 (2013.01); H04L 12/66 (2013.01); H04L 41/0894 (2022.05); H04L 41/145 (2013.01); H04L 41/5032 (2013.01); H04L 47/76 (2013.01); H04L 49/70 (2013.01); H04L 63/20 (2013.01); H04L 67/12 (2013.01); H04W 92/02 (2013.01); G06F 2009/45595 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system, comprising:
one or more processors; and
one or more computer-readable non-transitory storage media coupled to the one or more processors and comprising instructions that, when executed by the one or more processors, cause one or more switches of the system to perform operations comprising:
identifying a first end point using a protocol associated with the first end point;
determining a classification for the identified first end point based on one or more attributes of the first end point;
identifying one or more related end points having the classification in common with the first end point;
segmenting the first end point with the identified one or more related end points;
collecting telemetry data from the segmented first end point and the one or more related end points;
determining, using a deep learning model and the collected telemetry data from a particular segment associated with the segmented first end point and the one or more related end points, a segment-specific policy update to assist the segmented first end point and the one or more related end points in automatic troubleshooting, and
applying the policy update to the segmented first end point and the one or more related end points.