Dynamic Power Management for Blade Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern computing systems face oversubscription to power resources when multiple power supplies fail, leading to system shutdowns due to insufficient power delivery to blade components.
Innovation Solution
A power management module detects changes in power delivery capabilities and updates power settings for integrated technology elements based on a predefined redundancy policy, ensuring that power demand is aligned with available supply to prevent oversubscription.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If power supplies are shared among multiple blade components to improve resource utilization, then power efficiency is improved, but the risk of oversubscription increases when multiple power supplies fail simultaneously
Solution Approach 1:
The power management module dynamically adjusts power allocation and settings for blade components based on real-time detection of power supply status. When power delivery capabilities change, the system automatically updates power settings to prevent oversubscription, transforming a static power allocation system into a dynamic one that adapts to failure conditions.
Solution Approach 2:
The system implements a feedback mechanism where the power management module continuously monitors power supply status and detects changes in power delivery capabilities. Based on this feedback and according to a power redundancy policy, the module determines and applies appropriate power settings to maintain reliable operation under varying power availability.
2Reliability
If power settings are statically configured to ensure sufficient power headroom for failures, then reliability is improved, but power waste increases during normal operation
Solution Approach 1:
Instead of static power configuration, the system uses dynamic power management where settings are adjusted in real-time based on actual power supply status. During normal operation, power settings can be optimized for efficiency, while automatically adapting when failures occur to ensure adequate power availability.
Solution Approach 2:
The system changes power parameters (power settings for blade components) based on detected power supply capabilities. A power redundancy policy guides these parameter changes, allowing the system to maintain appropriate power headroom during failures while avoiding excessive power allocation during normal operation.
3Productivity
If the system allows oversubscription to maximize resource utilization, then productivity is improved, but system stability deteriorates when power demand exceeds supply
Solution Approach 1:
The system takes preliminary action by detecting changes in power delivery capabilities before oversubscription occurs. According to a power redundancy policy, the power management module proactively determines and applies appropriate power settings to prevent the harmful condition of oversubscription, rather than reacting after the problem arises.
Solution Approach 2:
The feedback mechanism enables the system to monitor power supply status and adjust power allocation accordingly. This continuous monitoring and adjustment prevents the system from entering an unstable oversubscribed state while still allowing high resource utilization during normal operation.
Data Source
AI summary
Preventing oversubscription to power resources in a computing system that includes a plurality of power supplies configured to deliver power to a plurality of integrated technology elements (‘ITEs’), including: detecting, by a power management module, that power delivery capabilities of the power supplies have changed; determining, by the power management module in dependence upon a power redundancy policy and the power delivery capabilities of the power supplies, power settings for the plurality of ITEs; and updating, by the power management module, power settings for the plurality of ITEs.