Data Center Thermal Model for Rack Capacity Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data center management tools lack the ability to accurately predict maximum cooler and rack capacities, as well as relevant temperatures, which are crucial for optimizing cooling performance and energy efficiency in data centers.

Innovation Solution

A computer-implemented method and system that evaluates data center equipment by receiving data on rack layouts and power draw, determining maximum cooler and rack capacities, and displaying these capacities, while also calculating cooling performance based on air flows and ambient temperatures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional data center management tools are used, then standardized design methodology is provided, but accurate prediction of maximum cooler and rack capacities and relevant temperatures is not achieved

Engineering Contradiction:
Improvetemperature prediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces conventional simplified thermal modeling methods with a physics-based thermal model that uses computational fluid dynamics principles to accurately predict temperature distributions, cooler capacities, and rack capacities in data centers, thereby achieving precise measurement without excessive system complexity

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a specialized thermal analysis system as an intermediary between conventional management tools and accurate temperature prediction requirements. This system acts as a mediator that bridges the gap by implementing sophisticated thermal modeling algorithms that can accurately predict capacities and temperatures while maintaining integration with existing data center management workflows

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of energy

If cooling performance optimization is pursued, then energy efficiency improves, but computational complexity and analysis time increase

Engineering Contradiction:
Improvecooling energy efficiencyVSAvoidanalysis time
Core Design Contradiction:
Loss of energyVSLoss of time

Solution Approach 1:

The patent performs preliminary thermal analysis and capacity determination during the design and planning phases, allowing optimization of cooling performance before actual deployment. By pre-calculating temperature distributions and capacities using the thermal model, the system enables energy efficiency optimization without requiring time-consuming analysis during operational phases

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements a dynamic thermal model that can adapt to different operating conditions and configurations. The model dynamically adjusts calculations based on input parameters such as rack layouts, power densities, and cooling configurations, providing accurate predictions without requiring exhaustive analysis for every scenario by leveraging learned patterns and simplified calculations for common configurations

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2580697B1System and method for predicting temperature values in a data center
Publication Date: 2018.11.14 SCHNEIDER ELECTRIC IT CORP
  • EP2580697B1 patent drawingFigure 1
  • EP2580697B1 patent drawingFigure 2
  • EP2580697B1 patent drawingFigure 3

AI summary

A system and method for evaluating equipment in a data center, the equipment including a plurality of equipment racks, and at least one cooling provider. In one aspect, a method includes receiving data regarding each of the plurality of equipment racks and the at least one cooling provider, the data including a layout of the equipment racks and the at least one cooling provider, and a power draw value for each of the equipment racks, storing the received data, determining air flow between the at least one cooling provider and each of the equipment racks, determining inlet and exit air temperature for the at least one cooling provider based on the layout, the power draw and the airflow, for each equipment rack, determining inlet and exit air temperature based on the layout, the power draw and the airflow, and displaying an indication of the inlet and exit temperature for each of the plurality of equipment racks and the at least one cooler. In the method, determining the inlet and exit temperature of each of the equipment racks and the at least one cooling provider includes establishing a set of S coupled equations, with S equal to a number of temperature values to be determined, and solving the S coupled equations.