PSU Degradation Detection via Telemetry Regression Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Power supply units (PSUs) in high-end computer servers often lack internal diagnostics and sensors, leading to unnoticed fan failures due to degradation, which can cause server shutdowns and increased downtime, necessitating costly manual replacements and potential wholesale recalls.
Innovation Solution
A system that analyzes telemetry data using nonlinear, nonparametric regression techniques to assess PSU integrity by comparing temperature and fan speed signals to reference values, performing remedial actions such as notifying technicians of degradation risks, and facilitating selective replacements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If commodity PSUs are used without internal diagnostics and sensors, then device complexity and manufacturing cost are reduced, but reliability and detectability of degradation are worsened
Solution Approach 1:
The patent introduces an intermediary monitoring system that uses existing system sensors and telemetry infrastructure to detect PSU degradation. Instead of adding sensors to the PSU itself, the system uses intermediaries (system-level temperature sensors, power management controllers) to indirectly monitor PSU health through parameters like temperature differentials, power delivery variations, and fan speed anomalies.
Solution Approach 2:
The patent replaces mechanical/direct sensing approaches with electronic/software-based monitoring. Instead of physical sensors inside the PSU, the system uses electronic telemetry data collection, analysis algorithms, and software-based diagnostic techniques to detect degradation, substituting mechanical complexity with electronic intelligence.
2Measurement precision
If manual investigation techniques are used to detect degraded PSUs, then measurement precision is improved, but loss of time and productivity are worsened
Solution Approach 1:
The patent implements continuous feedback monitoring where telemetry data from the PSU and system sensors is constantly collected, analyzed, and used to update degradation assessments. The system provides real-time feedback about PSU health status, enabling proactive identification of degradation trends before failure occurs, rather than relying on periodic manual checks.
Solution Approach 2:
The monitoring system performs self-diagnosis by automatically analyzing its own telemetry data to identify PSU degradation. The system serves itself by using its existing sensors and data infrastructure to monitor its own health, eliminating the need for external manual investigation while maintaining high detection precision.
3Reliability
If wholesale PSU replacement is performed during recalls, then reliability is improved, but loss of substance and productivity are worsened
Solution Approach 1:
The patent changes the decision parameter for PSU replacement from blanket replacement based on recall notices to targeted replacement based on actual degradation parameters. By monitoring specific parameters (temperature differentials, power delivery characteristics, fan performance), the system identifies which PSUs actually need replacement, changing the replacement criterion from categorical to parameter-based.
Solution Approach 2:
The system performs preliminary identification of degraded PSUs before failure occurs, allowing for planned, targeted replacement rather than reactive wholesale replacement. By detecting degradation early through continuous monitoring, the system enables proactive maintenance scheduling and precise identification of units requiring replacement, preventing unnecessary replacements.
Data Source
AI summary
Some embodiments provide a system that analyzes telemetry data from a computer system. During operation, the system obtains the telemetry data as a set of telemetric signals from the computer system and validates the telemetric signals using a nonlinear, nonparametric regression technique. Next, the system assesses the integrity of a power supply unit (PSU) in the computer system by comparing the telemetric signals to one or more reference telemetric signals associated with the computer system. If the assessed integrity falls below a threshold, the system performs a remedial action for the computer system.