Dynamic Hardware Power Cycling for Failure Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Complex computing systems face unexpected power-up failures after a power loss, leading to delayed system recovery due to simultaneous hardware failures detection during the power-up process, which is a significant issue for enterprise-level IT data centers.

Innovation Solution

A method involving dynamic power cycling of hardware components based on their age and expected lifespan, with periodic power cycling frequency increasing as the component approaches its lifespan, and routing I/O requests to redundant components during power cycling to ensure minimal disruption and detect potential failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If hardware components are power cycled frequently to detect potential failures, then system reliability is improved, but system productivity deteriorates due to increased downtime and operational disruptions

Engineering Contradiction:

Improvesystem reliabilityVSAvoidsystem productivity

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system implements periodic power cycling of hardware components at scheduled intervals rather than continuously or on-demand. This allows the system to maintain reliability through regular testing while minimizing productivity impact by confining disruptions to predetermined maintenance windows when system output is naturally reduced or stopped.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system performs power cycling as a preliminary preventive maintenance action before failures occur. By proactively testing components during scheduled maintenance windows, the system identifies potential failures early and replaces components before they cause unexpected downtime, thereby improving reliability without requiring reactive interruptions to productivity.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If power cycling is performed during system operation to maintain reliability, then hardware failures are detected early, but system availability deteriorates due to component unavailability during testing

Engineering Contradiction:

Improvehardware failure detectionVSAvoidsystem availability

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Power cycling is scheduled periodically during predetermined maintenance windows rather than continuously or during peak operation periods. This timing strategy ensures that components are tested for failures when system availability requirements are naturally lower, thereby maintaining reliability through regular testing while minimizing the impact on system availability.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system uses redundant hardware components to maintain operational capacity during power cycling of primary components. When a component is power-cycled for testing, its redundant counterpart can handle the workload, thereby detecting potential failures in the tested component without causing loss of system availability.

Inventive Principle:

Principle #26Copying

3Loss of time

If redundant components are used to maintain system operation during power cycling, then system availability is maintained, but device complexity increases due to additional hardware requirements

Engineering Contradiction:

Improvesystem availabilityVSAvoidhardware redundancy configuration

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The system schedules power cycling during predetermined maintenance windows when system availability requirements are naturally reduced. This timing approach allows the system to maintain availability during critical periods while performing necessary maintenance during lower-demand periods, thereby reducing the need for extensive redundant hardware while still maintaining adequate availability.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system dynamically manages component states during power cycling by temporarily transitioning components between active and maintenance modes. This dynamic management allows the system to optimize the use of existing redundant components only when needed, rather than requiring permanent redundant configurations for all components, thereby reducing overall device complexity while maintaining availability during critical operations.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11023029B2Preventing unexpected power-up failures of hardware components

Publication Date: 2021.06.01 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

In one embodiment, a method includes determining a plurality of hardware components of a system. The method also includes power cycling a first hardware component of the plurality of hardware components of the system according to a dynamic schedule. A period of time in which power cycling of the first hardware component takes place is shortened as the age of the first hardware component approaches the expected lifespan of the first hardware component. Also, the method includes determining whether the first hardware component experienced a power-up failure resulting from the power cycling. Moreover, the method includes outputting an indication to replace and/or repair the first hardware component in response to a determination that the first hardware component experienced the power-up failure resulting from the power cycling. Other systems, methods, ad computer program products for preventing unexpected power-up failures of individual hardware components are described in accordance with more embodiments.