Apparatus and method for system profile learning in an information handling system

Patent metadata

Field	Value
Publication number	US-11308417-B2
Application number	US-201715594012-A
Country	US
Kind code	B2
Filing date	May 12, 2017
Priority date	Aug 11, 2014
Publication date	Apr 19, 2022
Grant date	Apr 19, 2022

Abstract

An information handling system includes a processing system including a first sensor, and a second sensor, and a management system including an anomaly table, a learned model table entry associated with the processing system and including a learned model and a first sensor data history, and a prediction module to implement a prediction algorithm. The management system is configured to: receive first sensor data and second sensor data, determine an estimate of a first value of the first sensor data using a second value of the second sensor data, determine a residual of the first value by a comparison of the estimate to the first value, determine a significance of the residual, where the significance having a significant value is associated with a predicted anomaly, determine that an anomaly table entry has a known anomaly class for the predicted anomaly, and perform a remediation plan to resolve the predicted anomaly.

First claim

What is claimed is:

1. An information handling system, comprising:
a processing system including:
a first hardware sensor configured to provide first sensor data; and
a second hardware sensor configured to provide second sensor data; and

a management system coupled to the processing system and including:
an anomaly table;
a learned model table entry associated with the processing system and including a learned model and a first sensor data history; and
a prediction module to implement a prediction algorithm; and

wherein the management system is configured to:
receive the first sensor data and the second sensor data;
determine, based on the prediction algorithm and the learned model, an estimate of a first value of the first sensor data using a second value of the second sensor data;
determine a residual of the first value by a comparison of the estimate to the first value;
determine, based on the learned model and the first sensor data history, a significance of the residual, wherein the significance having a significant value is associated with a predicted anomaly;
determine that an anomaly table entry in the anomaly table has a known anomaly class for the predicted anomaly, based on the first value, the second value, and known anomaly classes in the anomaly table, in response to the significance having the significant value;
perform a remediation plan of the anomaly table entry to resolve the predicted anomaly in response to determining the known anomaly class, wherein the remediation plan includes at least one of: a heat-sink dust removal process;

an inlet vent clearing process; and a re-application of heat-sink thermal grease; and wherein the remediation plan further includes: a fan replacement; a replacement/balancing of laptop base feet; and a device replacement;
define a new anomaly class for the predicted anomaly in response to determining that the predicted anomaly has an unknown anomaly class;
add the new anomaly class to a new anomaly table entry in the anomaly table; and
add a severity and a remediation plan to the new anomaly table entry.

2. The information handling system of claim 1 , wherein the management system is further configured to:
update the learned model to account for at least one of an age of the processing system, a degradation in processing system performance, and a processing system wear-out condition.

3. The information handling system of claim 1 , wherein the first hardware sensor and the second hardware sensor includes at least one of a fan sensor, a power sensor, a thermal sensor, a performance sensor, and a system load sensor.

4. The information handling system of claim 1 , wherein the first sensor data and the second sensor data includes at least one of a fan speed, a fan operating state, a voltage level, a current level, a power supply operating state, a temperature of the processing system, the temperature of a device, an ambient temperature of the processing system, the ambient temperature of the device, a processor performance, a memory access performance, an input/output device performance, a processor utilization, a cache state residency, a memory utilization, a built in operating system (BIOS) log of the processing system, and a communication log between the processing system and a remote information handling system.

5. The information handling system of claim 1 , wherein the prediction algorithm includes a machine learning algorithm.

6. The information handling system of claim 5 , wherein the prediction algorithm further includes a Bayesian average, a weighted average, a basic average, a rule of succession, and other forms of additive smoothing.

7. The information handling system of claim 1 , wherein:
the predicted anomaly includes a clogged heat-sink;
the first hardware sensor includes a fan sensor configured to provide a fan speed of a fan;
the second hardware sensor includes a thermal sensor configured to provide a processor temperature of a processor;
the processing system further includes a performance sensor configured to provide a processor utilization of the processor;
the first sensor data history includes a fan sensor data history;
the prediction algorithm includes a multi-variant least square algorithm; and
the management system is further configured to:
receive the fan speed, the processor temperature, and the processor utilization; and
determine, based on the multi-variant least square algorithm and the learned model, the estimate of the fan speed using the processor temperature and the processor utilization;
wherein:
determining the significance further includes utilizing fan speed outlier information and fan speed threshold information from the fan sensor data history; and
determining that the anomaly table entry has the known anomaly class for the clogged heat-sink is further based on the fan speed, the processor temperature, the processor utilization, and the known anomaly classes in the anomaly table.

8. A method comprising:
providing, by a processing system of an information handling system, first sensor data from a first sensor of the processing system to a management system of the information handling system;
providing, by the processing system, second sensor data from a second sensor of the processing system to the management system;
determining, by a prediction module of the management system, based on a prediction algorithm and a learned model of a learned model table entry associated with the processing system, an estimate of a first value of the first sensor data using a second value of the second sensor data;
determining, by the prediction module, a residual of the first value by a comparison of the estimate to the first value;
determining, by the prediction module, a significance of the residual, based on the learned model and a first sensor data history of the learned model table entry, wherein the significance having a significant value is associated with a predicted anomaly;
determining, by the prediction module, that an anomaly table entry in an anomaly table has a known anomaly class for the predicted anomaly, based on the first value, the second value, and known anomaly classes in the anomaly table, in response to the significance having a significant value;
performing, by the prediction module, a remediation plan of the anomaly table entry to resolve the predicted anomaly in response to determining the known anomaly class, wherein the remediation plan includes at least one of: a heat-sink dust removal process; an inlet vent clearing process; and a re-application of heat-sink thermal grease; and wherein the remediation plan further includes; a fan replacement; a replacement/balancing of laptop base feet; and a device replacement; and
receiving, by the management system, third sensor data from a third sensor of the processing system, wherein the first sensor data includes a fan speed of a fan, the second sensor data includes a processor temperature of a processor, the third sensor data includes a processor utilization of the processor, and the predicted anomaly includes a clogged heat-sink and the remediation plan includes a heat-sink dust removal process.

9. The method of claim 8 , further comprising:
defining, by the prediction module, a new anomaly class for the predicted anomaly in response to determining that the predicted anomaly has an unknown anomaly class;
adding, by the prediction module, the new anomaly class to a new anomaly table entry in the anomaly table; and
adding, by the prediction module, a severity and a remediation plan to the new anomaly table entry.

10. The method of claim 8 , further comprising:
updating, by the prediction module, the learned model to accoun

Assignees

• Dell Products Lp

Inventors

• Vichare Nikhil M
• Ning Yan

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F11/3452

  Performance evaluation by statistical analysis · CPC title

• G06F11/0709

  in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems · CPC title

• G06F11/30

  Monitoring · CPC title

• G06F11/0751

  Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06) · CPC title