Dual-Loop Datacenter Cooling for Mixed-Power Server Components

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing datacenter cooling systems face inefficiencies in cooling high-power and low-power server components, with air cooling being insufficient for high-power components and liquid immersion cooling not effectively handling both types due to lower heat capacity.

Innovation Solution

A dual-mode cooling system utilizing separate cooling loops for high-power components with cold plate cooling and low-power components with immersion cooling, employing different coolants to enhance efficiency and reduce power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If air cooling is used for all server components, then the cooling system is simple to implement, but high-power components cannot be cooled sufficiently

Engineering Contradiction:
Improvecooling system implementation simplicityVSAvoidcooling effectiveness for high-power components
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The cooling system is divided into two distinct segments: air cooling for low-power components and liquid immersion cooling for high-power components. This segmentation allows each cooling method to be optimized for its specific application, with air cooling handling components that generate less heat and liquid cooling handling high-power components that require more intensive cooling, thereby resolving the contradiction between system simplicity and cooling effectiveness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different cooling methods are applied to different parts of the server system based on their specific thermal requirements. Low-power components receive air cooling while high-power components receive liquid immersion cooling. This local differentiation of cooling quality ensures that each component receives the appropriate level of cooling intensity, maintaining reliability without unnecessarily complicating the entire system.

Inventive Principle:
Principle #3Local quality

2Reliability

If liquid immersion cooling is used for all components, then cooling effectiveness is improved, but cooling efficiency decreases due to lower heat capacity of the coolant

Engineering Contradiction:
Improvecooling effectivenessVSAvoidcooling efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system segments the cooling load between two different cooling methods: liquid immersion cooling for high-power components that require high cooling effectiveness, and air cooling for low-power components where liquid cooling would be inefficient. This segmentation resolves the contradiction by applying liquid cooling only where its superior heat removal capability is necessary, rather than using it universally.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the cooling parameter (cooling method) based on the power density of different components. High-power components use liquid immersion cooling with high heat capacity coolant for maximum heat removal, while low-power components use air cooling for better efficiency. This parameter change strategy optimizes the balance between cooling effectiveness and efficiency across the entire system.

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If a single cooling loop is used for all components, then the system structure is simple, but cooling efficiency is reduced due to mismatched cooling requirements

Engineering Contradiction:
Improvecooling loop structureVSAvoidcooling efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The cooling system is segmented into multiple loops: a first cooling loop for high-power components using liquid immersion cooling, and a second cooling loop for low-power components using air cooling. This segmentation allows each loop to be optimized for its specific thermal load, improving overall cooling efficiency while maintaining reasonable structural complexity through modular design.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The cooling system achieves multi-functionality by using a single integrated system that can handle both high-power and low-power components through different cooling methods. The system universally cools all server components but adapts the cooling method based on component requirements, effectively combining the functions of what would traditionally require separate cooling systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

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

The dual-mode cooling system achieves heightened cooling efficiency and reduced power consumption by effectively cooling both high-power and low-power components, maintaining compatibility with existing datacenter hardware.

Implementation Method 1

cold plate cooling to cool high-power server components

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Implementation Method 2

liquid immersion cooling to cool low-power server components

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Implementation Method 3

heat exchangers to reject heat from the coolants

Methodology Applied
Scientific EffectHeat exchange: Heat Exchanger

Data Source

PatentUS12477699B2Dual-mode datacenter cooling system with cooling loops and heat exchangers
Publication Date: 2025.11.18 NVIDIA CORP
  • US12477699B2 patent drawing
  • US12477699B2 patent drawing
  • US12477699B2 patent drawing

AI summary

A system includes one or more first cooling loops to cool one or more first components within one or more servers having a first power density, and one or more second cooling loops to cool one or more second components within the one or more servers having a second power density. The system can flow first coolant to cold plates to cool high-power server components and flow second coolant to cool low-power server components by immersion cooling.