US 11,860,854 B2
Anomaly detection in relational databases
Mahsa Shateri, Los Angeles, CA (US); and Steven Conforti, West Hartford, CT (US)
Assigned to Cigna Intellectual Property, Inc., Wilmington, DE (US)
Filed by Cigna Intellectual Property, Inc., Wilmington, DE (US)
Filed on Oct. 10, 2022, as Appl. No. 17/962,955.
Application 17/962,955 is a continuation of application No. 17/121,827, filed on Dec. 15, 2020, granted, now 11,468,041.
Prior Publication US 2023/0031594 A1, Feb. 2, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/00 (2019.01); G06F 16/23 (2019.01); G06F 16/28 (2019.01)
CPC G06F 16/2365 (2019.01) [G06F 16/285 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory component;
at least one processor communicatively coupled to the memory component, the at least one processor configured to:
access a data object including a data set that has been collected over a time interval;
select, from the data set, a reference data point representing a first resource utilization of a server at a starting point of the time interval, wherein the first resource utilization includes memory resources allocated by the server to one or more services;
select a second data point from the data set, the second data point representing a second resource utilization of the server during a sample collection period following the time interval;
compute an angle representing a resource utilization growth rate between the reference data point and the second data point, wherein the resource utilization growth rate includes at least the first resource utilization corresponding to the reference data point and the second resource utilization corresponding to the second data point;
determine that the angle representing the resource utilization growth rate exceeds a threshold; and
in response to determining that the angle representing the resource utilization growth rate exceeds the threshold,
determine that at least one resource utilization anomaly occurred; and
trigger an anomaly detection operation comprising at least one of: (i) rebooting the server and (ii) reallocating resources of the server.