US 11,777,823 B1Granted

Metric anomaly detection across high-scale data

AssigneeAmazon Technologies, Inc.

InventorsEknath Venkataramani, Amit Jagannath Mhatre

Life of this patent

Estimating the term…

Loading patent…

Abstract

An anomaly detection service of a provider network may be used to efficiently monitor for metric anomalies across a large number of IoT devices using mandatory and optional values for metrics. A client may configure any number of mandatory and optional values for a metric to be collected from IoT devices of a fleet. The client may also configure one or more criteria to by used for evaluating the mandatory values (e.g., a threshold percentage such as 99%). When the service receives metric values for the metric, the service determines whether the values satisfy the criteria for the mandatory value. If not, then the service indicates an anomaly. The service may also determine if any values other than the mandatory and optional values are present. If not, then the service indicates an anomaly.

Claims20

Dependent claims

A system, comprising:

one or more processors and one or more memories, wherein the one or more memories have stored thereon instructions, which when executed by the one or more processors, cause the one or more processors to implement an anomaly detection service for a plurality of clients of a provider network, wherein the anomaly detection service is configured to, for a given client of the plurality of clients of the anomaly detection service:
for a given metric to be monitored for a plurality of edge devices of the client:
receive, from the client, a first set of values as mandatory values to be associated with the metric;
receive, from the client, predefined criteria associated with the first set of values;
receive, from the client, a second set of values as optional values to be associated with the metric;
assign the first set of values to the metric as mandatory values for the metric;
assign the second set of values to the metric as optional values for the metric;
receive, from an edge device of the plurality of edge devices, a plurality of values for the metric;
determine, based on the plurality of values received from the edge device for the metric and the predefined criteria, that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric;
determine that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and the optional values for the metric;
and
in response to the determination that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric and that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and the optional values for the metric, continue monitoring of the edge device for additional values for the metric without indicating an anomaly has been detected for the metric.

9 more elements

A method, comprising:

performing, by one or more computing devices of a provider network that implement an anomaly detection service:
receiving, from an edge device of a plurality of edge devices of a client, a plurality of values for a metric;
determining, based on the plurality of values received from the edge device for the metric and predefined criteria associated with mandatory values for the metric, that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric, wherein a first set of values is assigned to the metric as the mandatory values for the metric;
determining that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and optional values for the metric, wherein a second set of values is assigned to the metric as the optional values for the metric;
and
in response to determining that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric and that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and the optional values for the metric, continue monitoring the edge device for additional values for the metric without indicating an anomaly has been detected.

3 more elements

One or more non-transitory computer-accessible storage media storing program instructions that when executed on or across one or more processors cause the one or more processors to:

receive, from an edge device of a plurality of edge devices of a client, a plurality of values for a metric;
determine, based on the plurality of values received from the edge device for the metric and predefined criteria associated with mandatory values for the metric, that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric, wherein a first set of values is assigned to the metric as the mandatory values for the metric;
determine that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and optional values for the metric, wherein a second set of values is assigned to the metric as the optional values for the metric;
and
in response to the determination that the plurality of values received from the edge device satisfies the predefined criteria for the mandatory values for the metric and that the plurality of values received from the edge device do not include any values that are not present in the mandatory values for the metric and the optional values for the metric, continue monitoring of the edge device for additional values for the metric without indicating an anomaly has been detected.

2 more elements

Dependencies are inferred from the claim text.