Blade Power Management Controller for PSU Failure Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In traditional blade management systems, failure of a power supply unit leads to throttling of high-priority blades even when sufficient power capacity remains, resulting in decreased performance due to inefficient power distribution.

Innovation Solution

A method and system where a management controller determines the reduced power capacity of non-failed power supply units and powers down low-priority resources instead of high-priority ones, ensuring that high-priority blades continue to operate without throttling by assessing the aggregate minimum power requirements and available power capacity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the chassis management controller throttles all blades to minimum operating level in response to power supply unit failure, then as many blades as possible can receive power, but high-priority blades experience decreased performance

Engineering Contradiction:
Improvepower distribution reliabilityVSAvoidblade performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies local quality by differentiating power management strategies based on blade priority levels. High-priority blades receive uninterrupted power supply at full capacity, while low-priority blades are throttled to minimum levels or powered down. This localized differentiation ensures that critical services maintain optimal performance while non-critical services accept reduced performance, resolving the contradiction between reliable power distribution and maintaining blade productivity.

Inventive Principle:
Principle #3Local quality

2Reliability

If the chassis management controller powers off low-priority blades to maintain high-priority blade operation, then high-priority blades can continue operating, but high-priority blades may be unnecessarily throttled when sufficient power capacity exists

Engineering Contradiction:
Improvehigh-priority blade operationVSAvoidhigh-priority blade performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent implements feedback mechanisms where the chassis management controller continuously monitors available power capacity from remaining power supply units and adjusts power distribution accordingly. When sufficient power capacity exists, the system maintains high-priority blades at full operation without throttling. When power capacity becomes insufficient, the system selectively powers off low-priority blades. This feedback-driven adaptive approach prevents unnecessary throttling of high-priority blades while ensuring their operation during power failures.

Inventive Principle:
Principle #23Feedback

3Device complexity

If traditional approaches throttle all blades to minimum level, then power distribution is simplified, but aggregate power capacity becomes insufficient for all blades

Engineering Contradiction:
Improvepower management complexityVSAvoidaggregate power capacity
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent resolves this contradiction by implementing local quality differentiation in power management. Instead of uniformly throttling all blades, the system applies different power levels to different blade groups: high-priority blades receive full power capacity while low-priority blades receive minimum or zero power. This differentiated approach maximizes aggregate power capacity utilization and ensures sufficient power remains for critical services, while the management complexity is contained through priority-based decision logic.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS8006112B2System and method for managing blades after a power supply unit failure
Publication Date: 2011.08.23 DELL PROD LP
  • US8006112B2 patent drawing
  • US8006112B2 patent drawing
  • US8006112B2 patent drawing

AI summary

Systems and methods for managing blades in the event of a power supply unit failure are disclosed. A method may include determining whether a reduced power capacity of the non-failed power supply units is sufficient to provide an aggregate minimum power requirement of the resources in response to a failure of a particular power supply unit. The method may also include powering down a low-priority resource and not powering down a second resource having a higher priority than the low-priority resource if the reduced power capacity of the non-failed power supply units is not sufficient to provide the aggregate minimum power requirement of the resources.