Multi-Channel Processor Load Shedding Under Cooling Failure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current multi-channel processing systems face challenges in managing heat dissipation, particularly in high-power applications, where passive conduction cooling is insufficient, and external active cooling mechanisms like fans offer limited reliability, leading to common mode failures when cooling losses occur, causing entire channels to be disabled due to non-prioritized load shedding.

Innovation Solution

A distributed multi-channel processing system dynamically prioritizes the shedding of dissimilar processors across channels, reducing processor functionality based on cooling element failures, and exchanging status information to prevent common mode failures by disabling processors in a common mode tolerant manner, ensuring continued operation within operational temperature ranges.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Temperature

If external active cooling mechanisms like fans are used to manage heat dissipation in high-power applications, then cooling effectiveness is improved, but reliability deteriorates due to limited reliability of cooling components causing common mode failures

Engineering Contradiction:
Improveheat dissipationVSAvoidcooling system reliability
Core Design Contradiction:
TemperatureVSReliability

Solution Approach 1:

The system segments the processing channels into multiple independent groups, each with its own cooling resources. When a cooling failure occurs in one segment, only that segment is affected while other segments continue operating independently, preventing system-wide common mode failures.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system pre-configures multiple cooling resources and establishes cooling relationships before failures occur. When a cooling element fails, the system can immediately activate pre-arranged backup cooling resources without delay, maintaining reliability through proactive preparation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If redundant processors are added to ensure system reliability, then fault tolerance is improved, but heat generation increases requiring additional cooling resources

Engineering Contradiction:
Improvesystem reliabilityVSAvoidheat generation
Core Design Contradiction:
ReliabilityVSTemperature

Solution Approach 1:

The system assigns different cooling resources to different processor types within channels, creating localized cooling solutions tailored to specific processor requirements. This allows efficient heat management for each processor type without requiring uniform over-cooling of the entire system.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts cooling resource allocation based on operational conditions and processor load. When processors are disabled due to cooling failures, the system changes operational parameters to maintain reliability with reduced heat generation from active components.

Inventive Principle:
Principle #35Parameter changes

3Temperature

If cooling resources are reduced to lower heat generation, then temperature control is improved, but cooling effectiveness deteriorates when cooling losses occur

Engineering Contradiction:
Improvetemperature controlVSAvoidcooling effectiveness
Core Design Contradiction:
TemperatureVSReliability

Solution Approach 1:

The system pre-establishes multiple cooling relationships and backup cooling resources before failures occur. When a cooling element fails, previously configured backup resources are immediately activated, providing cushioning against the failure and maintaining cooling effectiveness without requiring continuous high-power cooling operation.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

4Reliability

If dissimilar processors are used in each channel to maximize tolerance to common mode failures, then fault tolerance is improved, but system complexity increases

Engineering Contradiction:
Improvefault toleranceVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system divides processors into distinct channels with dissimilar processor types, segmenting the complexity management. Each channel can be independently configured and managed, making the overall complex system tractable through modular organization of dissimilar components.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances reliability by maximizing redundancy across dissimilar processor types, preventing common mode failures and maintaining system operation in mission-critical applications like flight and engine control systems, even with multiple cooling losses.

Implementation Method 1

External active cooling commonly uses fans to move air across a chassis

Methodology Applied
Scientific EffectForced Convection: Forced Convection

Implementation Method 2

passive conduction cooling is insufficient

Methodology Applied
Scientific EffectThermal Conduction: Conduction (thermal)

Data Source

PatentEP3594780B1Intelligent load shedding for multi-channel processing systems
Publication Date: 2021.06.16 HAMILTON SUNDSTRAND CORP
  • EP3594780B1 patent drawingFigure 1
  • EP3594780B1 patent drawingFigure 2
  • EP3594780B1 patent drawingFigure 3

AI summary

Embodiments in include a system, a method, and a computer program product for performing intelligent load shedding for multi-channel processing system. The embodiments include a multi-channel processing system, wherein each channel of the multi-channel processing system includes a plurality of processors (1, 2, 3), and a plurality of links coupling each channel with each other channel in the multi-channel processing system, wherein the links are used to transmit status information of the plurality of processors. The embodiments also include a plurality of cooling elements (210, 212, 214) coupled to each channel having the plurality of processors, wherein the plurality of cooling elements (210, 212, 214) are configured to remove heat from the multi-channel processing system.