Back to results
US 11,237,898 B2Granted
Automatic model-based computing environment performance monitoring
AssigneeIntel Corporation
InventorsGiovani Estrada, Victor Bayon-Molino
Life of this patent
Estimating the term…
Abstract
Various systems and methods for implementing automatic model generation for performance monitoring are described herein. A performance monitoring system includes a model manager to: identify a performance model that predicts performance of an operational node, the performance model based on telemetry data from the operational node; and implement an automatic verification operation to analyze the performance model and revise the performance model when the performance model is no longer valid; and an event processor to: initiate a remedial action at the operational node when the performance model indicates an alert state.
Claims24
An edge computing node comprising:
- at least one processor;
- and
- memory including instructions that, when executed by the at least one processor, cause the at least one processor to perform operations to:
- generate a plurality of quality service models based on training telemetry data for an application executing on the edge computing node, the plurality of quality service models used to evaluate performance metrics of the edge computing node to predict an operational state of the application;
- deploy the plurality of quality of service models to the application executing on the edge computing node;
- apply an adaptation to the application executing on the edge computing node via a monitoring entry point of the application executing on the edge computing node to meet a service level agreement, wherein the adaptation corresponds to a quality of service model selected from the plurality of quality of service models deployed within the application based on an initial evaluation of telemetry data of the edge computing node;
- initiate a remedial action via the adapted application executing on the edge computing node in response to identification of a variation from the service level agreement based on evaluation of the telemetry data evaluated with the quality of service model;
- and
- invoke an automatic verification operation to analyze the quality of service model and revise the quality of service model and the adaptation in response to identifying that updated telemetry data indicates non-conformance to the service level agreement.
6 more elements
At least one non-transitory machine-readable medium including instructions for performance monitoring at an edge computing node that, when executed by at least one processor, cause the at least one processor to perform operations to:
- generate a plurality of quality service models based on training telemetry data for an application executing on the edge computing node, the plurality of quality service models used to evaluate performance metrics of the edge computing node to predict an operational state of the application;
- deploy the plurality of quality of service models to the application executing on the edge computing node;
- apply an adaptation to the application executing on the edge computing node via a monitoring entry point of the application executing on the edge computing node to meet a service level agreement, wherein the adaptation corresponds to a quality of service model selected from the plurality of quality of service models deployed within the application based on an initial evaluation of telemetry data of the edge computing node;
- initiate a remedial action at the edge computing node via the adapted application executing on the edge computing node in response to identification of a variation from the service level agreement based on evaluation of the telemetry data evaluated with the quality of service model;
- and
- invoke an automatic verification operation to analyze the quality of service model and revise the quality of service model and the adaptation in response to identifying that updated telemetry data indicates non-conformance to the service level agreement.
3 more elements
A method for performance monitoring at an edge computing node comprising:
- generating a plurality of quality service models based on training telemetry data for an application executing on the edge computing node, the plurality of quality service models used to evaluate performance metrics of the edge computing node to predict an operational state of the application;
- deploying the plurality of quality of service models to the application executing on the edge computing node;
- applying an adaptation to the application executing on the edge computing node via a monitoring entry point of the application executing on the edge computing node to meet a service level agreement, wherein the adaptation corresponds to a quality of service model selected from the plurality of quality of service models deployed within the application based on an initial evaluation of telemetry data of the edge computing node;
- initiating a remedial action at the edge computing node via the adapted application executing on the edge computing node in response to identification of a variation from the service level agreement based on evaluation of the telemetry data evaluated with the quality of service model;
- and
- invoking an automatic verification operation to analyze the quality of service model and revise the quality of service model and the adaptation in response to identifying that updated telemetry data indicates non-conformance to the service level agreement.
3 more elements
Dependencies are inferred from the claim text.