Hardware-based system monitoring
An exemplary monitoring system receives log data associated with an operation of a hardware component, applies the log data as an input to an unsupervised machine learning model, and identifies, based on an output of the unsupervised machine learning model, an anomaly in the log data.
1. A method comprising:
receiving, by a monitoring system, log data associated with an operation of a hardware component;
applying, by the monitoring system, the log data as an input to an unsupervised machine learning model;
identifying, by the monitoring system based on an output of the unsupervised machine learning model, an anomaly in the log data;
determining, by the monitoring system, that the anomaly is representative of an issue associated with the hardware component; and
performing, by the monitoring system based on the determining that the anomaly is representative of the issue associated with the hardware component, a remedial action associated with the anomaly by performing one or more of
slowing down a performance of the operation of the hardware component,
preventing the operation of the hardware component from being performed,
disabling the hardware component, or
directing an additional hardware component to perform the operation in place of the hardware component.
2. The method of claim 1 , wherein the determining that the anomaly is representative of the issue associated with the hardware component comprises applying data representative of the anomaly as an input to a supervised machine learning model, the supervised machine learning model configured to:
determine a confidence score for the anomaly;
determine that the confidence score is above a threshold associated with the issue; and
classify, in response to the determination that the confidence score is above the threshold, the anomaly as being representative of the issue associated with the hardware component.
3. The method of claim 2 , further comprising:
receiving, by the monitoring system, user input confirming or refuting that the anomaly is an issue associated with the hardware component; and
providing, by the monitoring system, the user input as a training input to the supervised machine learning model.
4. The method of claim 1 , wherein the determining that the anomaly is representative of the issue associated with the hardware component comprises:
applying a rule set to data representative of the anomaly to generate a confidence score for the anomaly; and
determining that the confidence score is above a threshold associated with the issue.
5. The method of claim 1 , further comprising:
identifying, by the monitoring system based on an output of the unsupervised machine learning model, an additional anomaly in the log data;
determining, by the monitoring system, that the additional anomaly is not representative of an issue associated with the hardware component; and
abstaining, by the monitoring system in response to the determining that the additional anomaly is not representative of the issue associated with the hardware component, from performing a remedial action associated with the additional anomaly.
6. The method of claim 1 , wherein the receiving of the log data comprises:
receiving, by way of a network, phone-home logs from a hardware-based system that includes the hardware component; and
extracting the log data from the phone-home logs.
7. The method of claim 1 , wherein the receiving of the log data comprises extracting the log data with an application executed by a hardware-based system that includes the hardware component.
8. The method of claim 1 , wherein the log data comprises at least one of data associated with a physical aspect of the hardware component and data associated with an operation of a software component that utilizes the hardware component.
9. The method of claim 1 , wherein the log data comprises at least one of data representative of a power consumption of the hardware component, data indicating an occurrence of a power recycle of the hardware component, data representative of an alert generated with respect to the hardware component, data representative of an occurrence of an event associated with the hardware component, data associated with a wear cycle of the hardware component, data representative of a temperature of the hardware component, data representative of a clock speed of the hardware component, data representative of a usage of the hardware component, and data representative of an available storage capacity of the hardware component.
10. The method of claim 1 , wherein the unsupervised machine learning model is configured to implement a variational autoencoder heuristic by:
encoding the log data to generate encoded data;
decoding the encoded data to generate decoded data;
determining an error measurement that represents a decoding error between the log data and the decoded data; and
generating, based on the decoding error, a confidence score for a data subset of the log data that indicates that the data subset includes the anomaly.
11. The method of claim 1 , further comprising identifying, by the monitoring system based on the anomaly, a performance characteristic of the hardware component.
12. A system comprising:
a memory storing instructions;
a processor communicatively coupled to the memory and configured to execute the instructions to:
receive log data associated with an operation of a hardware component,
apply the log data as an input to an unsupervised machine learning model,
use an output of the unsupervised machine learning model to identify an anomaly in the log data;
determine that the anomaly is representative of an issue associated with the hardware component; and
perform, based on the determining that the anomaly is representative of the issue associated with the hardware component, a remedial action associated with the anomaly by performing one or more of
slowing down a performance of the operation of the hardware component,
preventing the operation of the hardware component from being performed,
disabling the hardware component, or
directing an additional hardware component to perform the operation in place of the hardware component.
13. The system of claim 12 , wherein the determination that the anomaly is representative of the issue associated with the hardware component comprises applying data representative of the anomaly as an input to a supervised machine learning model, the supervised machine learning model configured to:
determine a confidence score for the anomaly;
determine that the confidence score is above a threshold associated with the issue; and
classify, in response to the determination that the confidence score is above the threshold, the anomaly as being representative of the issue associated with the hardware component.
14. The system of claim 12 , wherein the determining that the anomaly is representative of the issue associated with the hardware component comprises applying data representative of the anomaly as an input to a supervised machine learning model, the supervised machine learning model configured to:
determine a confidence score for the anomaly;
determine that the confidence score is above a threshold associated with the issue; and
classify, in response to the determination that the confidence score is above the threshold, the anomaly as being representative of the issue associated with the hardware component.
15. The system of claim 14 , wherein the processor is further configured to execute the instructions to:
receive user input confirming or refuting that the anomaly is an actual issue associated with the hardware component; and
provide the user input as a training input to the supervised machine learning model.
16. The system of claim 12 , wherein the determining that the anomaly is representative of the issue associated with the hardware component comprises:
applying a rule set to data representative of the anomaly to generate a confidence score for the anomaly; and
determining that the confidence score is above a threshold associated with the issue.
17. The system of claim 12 , wherein the processor is further configured to execute the instructions to:
identify, based on an output of the unsupervised machine learning model, an additional anomaly in the log data;
determine that the additional anomaly is not representative of an issue associated with the hardware component; and
abstain, in response to the determining that the additional anomaly is not representative of the issue associated with the hardware component, from performing a remedial action associated with the additional anomaly.
18. The system of claim 12 , wherein the receiving of the log data comprises:
receiving, by way of a network, phone-home logs from a hardware-based system that includes the hardware component; and
extracting the log data from the phone-home logs.
19. The system of claim 12 , wherein the receiving of the log data comprises extracting the log data with an application executed by a hardware-based system that includes the hardware component.
20. A non-transitory computer-readable medium storing instructions that, when executed, direct a processor of a computing device to:
receive log data associated with an operation of a hardware component;
apply the log data as an input to an unsupervised machine learning model;
use an output of the unsupervised machine learning model to identify an anomaly in the log data;
determine that the anomaly is representative of an issue associated with the hardware component; and
perform, based on the determining that the anomaly is representative of the issue associated with the hardware component, a remedial action associated with the anomaly by performing one or more of
slowing down a performance of the operation of the hardware component,
preventing the operation of the hardware component from being performed,
disabling the hardware component, or
directing an additional hardware component to perform the operation in place of the hardware component.