US 12,169,709 B2
Contextually cognitive edge server manager
Tiberiu Suto, Franklin, NY (US); Shikhar Kwatra, San Jose, CA (US); Raghuveer Prasad Nagar, Kota (IN); and Jagadesh Ramaswamy Hulugundi, Bangalore (IN)
Assigned to Kyndryl, Inc., New York, NY (US)
Filed by Kyndryl, Inc., New York, NY (US)
Filed on Mar. 30, 2022, as Appl. No. 17/708,405.
Prior Publication US 2023/0315438 A1, Oct. 5, 2023
Int. Cl. G06F 8/658 (2018.01); G06F 8/71 (2018.01); G06F 9/46 (2006.01); G06F 9/50 (2006.01); G06F 11/34 (2006.01); G06F 21/64 (2013.01); G06N 20/20 (2019.01)
CPC G06F 8/658 (2018.02) [G06F 8/71 (2013.01); G06F 9/466 (2013.01); G06F 9/5077 (2013.01); G06F 11/3457 (2013.01); G06F 21/64 (2013.01); G06N 20/20 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method performed by an edge manager device in a server cluster, the method comprising:
predicting a jeopardy associated with an edge server included in the server cluster that communicates with a central server;
responsive to the predicting the jeopardy associated with the edge server, detaching the edge server from the server cluster;
determining a patch based on the jeopardy associated with the edge server;
determining a software configuration and versioning information of the edge server;
generating a token using a hash function based on the software configuration and versioning information of the edge server;
encrypting the patch with the token as a password;
pushing the patch to the edge server;
validating the patch by performing a regression testing on the edge server after the patch has been applied to the edge server; and
inducting the edge server into the server cluster, wherein the inducting is performed in response to the validating the patch.