Hardware-based predictive fault detection and analysis
AssigneeHewlett Packard Enterprise Development LP
InventorsSuresh K. Thapa, Garnett H. Thompson, Abhishek C. Mehta
Life of this patent
Estimating the term…
Abstract
A method and system for hardware-based predictive fault detection and analysis are described herein. Logic components of a computing cluster can baseline a plurality of telemetries associated with at least one processing node of the computing cluster. The logic components can monitor the plurality of telemetries while the at least one processing node is in operation. The logic components of the computing cluster can compare the monitored plurality of telemetries with the baselined plurality of telemetries. The logic components can predict one or more impending faults associated with the at least one processing node based on the comparisons.
Claims20
A method for hardware-based predictive fault detection and analysis, the method comprising:
- baselining, by logic components of a computing cluster, a plurality of telemetries associated with at least one processing node of the computing cluster;
- monitoring, by the logic components, the plurality of telemetries while the at least one processing node is in operation;
- comparing, by the logic components, the monitored plurality of telemetries with the baselined plurality of telemetries;
- and
- predicting, by the logic components, one or more impending faults associated with the at least one processing node based on the comparisons.
A system comprising:
- at least one computing cluster comprising logic components, wherein the logic components are configured to perform a method for hardware-based predictive fault detection and analysis, the method comprising:
- baselining a plurality of telemetries associated with at least one processing node of the computing cluster;
- monitoring the plurality of telemetries while the at least one processing node is in operation;
- comparing the monitored plurality of telemetries with the baselined plurality of telemetries;
- and
- predicting one or more impending faults associated with the at least one processing node based on the comparisons.
Dependencies are inferred from the claim text.