Fault-Tolerant Power Control in Server Enclosures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power management systems for groups of computers, such as server enclosures, lack fault-tolerance and are not suitable for mission-critical environments as they rely on a single power controller for power control, which is not reliable.
Innovation Solution
A power subsystem with dual zones, each equipped with AC and DC power meters and controllers, allows for continuous power measurement and failover detection, enabling accurate power management and redundancy to ensure uninterrupted operation even if a single component fails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single power controller is used to control power, then device complexity is reduced, but reliability deteriorates because the system is not fault-tolerant
Solution Approach 1:
The power controller is divided into a primary power controller and a secondary power controller, each capable of independently controlling power to different zones of the enclosure. This segmentation allows the system to maintain fault-tolerance while managing complexity through clear division of responsibilities between controllers.
Solution Approach 2:
The system performs preliminary actions by having both primary and secondary controllers continuously monitor each other's health status and readiness. Failover detection mechanisms are pre-configured to detect controller failures before they impact power delivery, enabling seamless transition without service interruption.
2Reliability
If dual-zone redundancy is implemented, then reliability improves through fault-tolerance, but device complexity increases due to multiple controllers and meters
Solution Approach 1:
The primary and secondary power controllers are merged into a coordinated system where both controllers can perform identical functions. Each controller has redundant AC power meters and DC power meters, allowing either controller to independently measure and control power if the other fails, thus improving reliability while managing complexity through functional equivalence.
Solution Approach 2:
The system dynamically changes operational parameters by switching between primary and secondary controllers based on detected failure conditions. When a controller or its associated meters fail, the system changes the active controller parameter and adjusts power measurement sources accordingly, maintaining continuous operation without manual intervention.
3Measurement precision
If power measurement is performed by multiple meters, then measurement precision is maintained during failures, but device complexity increases
Solution Approach 1:
Each power controller is equipped with duplicate AC power meters and DC power meters that mirror each other's functionality. The secondary controller has identical measurement capabilities as the primary controller, creating redundant copies that can take over measurement functions if the primary meters fail, ensuring measurement precision is maintained during failures.
Data Source
AI summary
Fault-tolerant power control in a computer system is described. In an example, a power controller to control power applied to an enclosure having a plurality of computing devices includes: first and second alternating current (AC) primary power meters respectively measuring first and second input power feeds to the enclosure; a first alternative power meter to measuring power supplied by the first input power feed; a second alternative power meter to measure power supplied by the second input power feed; a first controller coupled to the first primary power meter and the second alternative power meter; and a second controller coupled to the second primary power meter and the first alternative power meter; wherein the first controller is coupled to the second controller.