Wireless Mesh CRAC Control for Data Center Overcooling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional data center cooling systems are inefficient, as they often overcool entire rooms to meet the environmental requirements of a single computing device, leading to significant energy wastage and high operational costs, and lack the ability to distinguish between the varying cooling needs of different electronic devices within the data center.

Innovation Solution

A wireless mesh network system that uses distributed sensors to monitor environmental conditions and adaptively control the volume and temperature of cooling air or liquid, allowing for precise management of cooling units through a distributed control architecture that includes wireless pressure differential sensors and temperature sensors, enabling coordinated control of multiple cooling units for optimal cooling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional cooling systems cool entire rooms to meet environmental requirements, then all computing devices are maintained within safe temperature ranges, but significant energy is wasted on overcooling areas that do not require such intensive cooling

Engineering Contradiction:
Improvetemperature control reliabilityVSAvoidcooling energy waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The data center cooling system is segmented into multiple independently controllable cooling zones, each with its own CRAC unit. This allows selective cooling of specific racks or areas based on actual thermal loads, rather than cooling the entire room uniformly. The segmentation enables precise targeting of cooling resources to where they are actually needed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the data center are provided with different cooling qualities and intensities based on local requirements. High-density computing racks receive more intensive cooling, while low-density areas receive minimal or no cooling. This local differentiation eliminates the energy waste of overcooling areas that do not require intensive cooling.

Inventive Principle:
Principle #3Local quality

2Reliability

If cooling air is pushed through all racks at high volume, then all computing devices receive adequate cooling, but the energy cost of moving the air increases significantly

Engineering Contradiction:
Improvecooling adequacyVSAvoidair movement energy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The cooling system dynamically adjusts air flow volumes based on real-time thermal conditions and workload distributions. Air flow rates are varied continuously rather than maintained at constant high levels, allowing the system to match cooling capacity with actual demand and reduce energy consumption during periods of lower thermal loads.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses thermal sensors and workload monitoring to automatically determine cooling requirements, eliminating the need for manual intervention. The cooling infrastructure serves itself by responding to actual thermal conditions, adjusting air flow volumes optimally without human input while maintaining adequate cooling where needed.

Inventive Principle:
Principle #25Self-service

3Adaptability or versatility

If a centralized control system is used to manage cooling, then coordination between multiple CRAC units can be achieved, but the complexity of the control architecture increases

Engineering Contradiction:
Improvecooling coordination capabilityVSAvoidcontrol architecture complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The centralized control system is segmented into multiple distributed control modules, each managing a specific cooling zone or CRAC unit. These modular control elements communicate with each other and with a central coordinator, allowing complex coordination tasks to be broken down into manageable local decisions, reducing overall system complexity while maintaining adaptability.

Inventive Principle:
Principle #1Segmentation

4Loss of energy

If precise control of cooling volume and temperature is implemented, then energy efficiency is improved, but the measurement and control precision requirements increase

Engineering Contradiction:
Improvecooling energy efficiencyVSAvoidtemperature and flow measurement accuracy
Core Design Contradiction:
Loss of energyVSMeasurement precision

Solution Approach 1:

The system implements continuous feedback loops using thermal sensors, flow meters, and workload monitors to measure actual cooling conditions. This feedback information is used to automatically adjust CRAC unit operations, maintaining optimal energy efficiency. The feedback mechanism compensates for measurement uncertainties by continuously adapting to actual conditions rather than relying solely on precise predetermined settings.

Inventive Principle:
Principle #23Feedback

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 reduces energy consumption by allowing for more precise control of cooling resources, optimizing energy usage, and ensuring that each computing device receives the appropriate amount of cooling based on its specific needs, thereby minimizing unnecessary energy expenditure.

Implementation Method 1

the temperatures of air supplied by one or more computer room air conditioning (CRAC) units are also detected

Methodology Applied
Scientific EffectTemperature sensing:

Implementation Method 2

A distributed array of wireless pressure differential sensors can be deployed to characterize the volume of air that is being distributed

Methodology Applied
Scientific EffectPressure differential measurement:

Implementation Method 3

cooling units, such as computer room air conditioning (A/C) or air handling units distribute cold air or cold liquid (such as water and air) to different racks via aisles between the racks

Methodology Applied
Scientific EffectAir conditioning cooling:

Data Source

PatentEP2895798B1Apparatus and method for controlling computer room air conditioning units (CRACS) in data centers
Publication Date: 2019.12.04 SYNAPSENSE CORP
  • EP2895798B1 patent drawingFigure 1
  • EP2895798B1 patent drawingFigure 2
  • EP2895798B1 patent drawingFigure 3

AI summary

Various embodiments provide an apparatus and method for controlling computer room air conditioning units (CRACs) in data centers. An example embodiment includes: receiving an alert message from a reporting wireless network sensor of a plurality of wireless network sensors via a wireless sensor network, the alert message including information indicative of a modification needed to an environmental condition, the alert message including an indication of stability or instability of a computer room air conditioning unit (CRAC) in a data center; and using the information indicative of a modification needed to an environmental condition at a networked controller to command a device capable of modifying the environmental condition to modify the environmental condition in a manner corresponding to the information in the alert message, commanding the device including controlling fan speed in the CRAC to correct an indication of instability.