Staggered Power Transition for Data Center Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data centers face challenges in maintaining uninterrupted power supply to computing devices, particularly during faults in the primary AC power source, which can lead to disruptions in data processing and storage operations.

Innovation Solution

A power distribution architecture that transitions computing devices from a secondary power source to a primary power source in a staggered manner, using a unique seed value for each device to determine a delay time, ensuring seamless power transfer and minimizing disruptions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all computing devices transition from secondary power source to primary power source simultaneously, then power restoration is achieved quickly, but power surges and system instability occur

Engineering Contradiction:
Improvepower supply stabilityVSAvoidpower surge
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the simultaneous power transition of multiple computing devices into staggered, individual transitions. Each computing device applies a unique random delay time before transitioning from secondary to primary power source, dividing the bulk transition into many small, distributed transitions that prevent power surge accumulation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by having each computing device calculate and apply a random delay time before power transition. This preliminary timing adjustment ensures that devices do not transition simultaneously, proactively preventing power surge issues before they occur.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If a random delay time is applied before power transition, then power surge is reduced, but transition time is extended

Engineering Contradiction:
Improvepower supply stabilityVSAvoidpower transition time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies partial action by using random delay times that are sufficient to stagger transitions and prevent power surge, but not excessively long to cause unnecessary downtime. The delay is calibrated to achieve the minimum necessary staggering effect without extending transition time more than needed.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If computing devices are powered by secondary power source during primary power fault, then uninterrupted power supply is maintained, but secondary power source capacity is depleted

Engineering Contradiction:
Improveuninterrupted power supplyVSAvoidbattery power
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements skipping by rapidly transitioning computing devices from secondary to primary power source as soon as primary power is restored. The random delay mechanism ensures quick, staggered transitions that minimize the duration secondary power source must sustain the load, reducing energy depletion while maintaining uninterrupted supply.

Inventive Principle:
Principle #21Skipping (Rushing through)

Data Source

PatentEP2038717B1Distributed power-up
Publication Date: 2019.04.10 GOOGLE LLC
  • EP2038717B1 patent drawingFigure 1
  • EP2038717B1 patent drawingFigure 2
  • EP2038717B1 patent drawingFigure 3

AI summary

A method for powering a system is described. The method includes receiving a signal that indicates availability of a primary power source to supply operating power to a plurality of computing devices, and responsive to the received signal, transitioning each of the plurality of computing devices from a secondary power source to receiving power from the primary power source after a delay time that is a function of a substantially unique seed value for each computing device.