Double-Mass Server Model for Transient Thermal Characterization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data center management systems fail to accurately characterize and utilize the idealized thermal mass of servers, leading to inefficient cooling designs and excessive energy usage, as they neglect the thermal properties of servers in transient scenarios, resulting in overly conservative cooling infrastructure and potential temperature prediction errors.
Innovation Solution
A computer-implemented method and system using a double-mass server model to determine the effective and exhaust temperatures of servers based on measured steady-state parameters and thermal properties, allowing for optimized data center design by adjusting server layouts to improve cooling efficiency and energy usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data center management systems are used that neglect thermal properties of servers, then the design methodology is standardized and predictable, but the cooling design becomes overly conservative and energy usage increases
Solution Approach 1:
The patent changes the thermal parameters from static to dynamic by implementing transient thermal models that account for time-varying heat generation and cooling conditions. This allows the system to adapt cooling designs to actual thermal behavior rather than using fixed conservative estimates, reducing energy consumption while maintaining reliability.
Solution Approach 2:
The patent introduces dynamic thermal modeling that captures transient thermal responses of servers during startup, shutdown, and load changes. This dynamic approach replaces static conservative designs with adaptive models that reflect actual operational conditions, enabling more efficient cooling strategies.
2Device complexity
If conventional cooling designs are used that neglect server thermal mass, then the design process is simple, but temperature prediction accuracy deteriorates
Solution Approach 1:
The patent transforms the thermal model from steady-state to transient by incorporating time-dependent parameters including thermal mass, heat capacity, and time-varying heat generation rates. This enables accurate prediction of temperature dynamics during transient operations while maintaining manageable model complexity through systematic parameter organization.
Solution Approach 2:
The patent segments the server thermal system into distinct thermal zones and components (processors, memory, chassis, cooling channels) with specific thermal properties. This segmentation allows accurate temperature prediction for each component while keeping the overall model structured and manageable.
3Reliability
If conservative cooling infrastructure is designed without considering actual thermal behavior, then reliability is maintained, but cooling efficiency decreases
Solution Approach 1:
The patent implements feedback mechanisms that use measured temperature data and thermal model predictions to continuously optimize cooling system operation. This feedback loop enables the system to maintain reliability by detecting thermal anomalies while improving efficiency by adjusting cooling capacity to match actual thermal demands rather than operating at fixed conservative levels.
Solution Approach 2:
The patent enables dynamic adjustment of cooling strategies based on real-time thermal conditions and predicted transient behavior. This allows the cooling system to adapt its operation to match actual server thermal loads, maintaining reliability through proactive thermal management while improving efficiency by avoiding unnecessary cooling capacity.
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 method enables more accurate prediction of temperature changes and efficient cooling performance by characterizing the idealized thermal mass of servers, optimizing data center layouts to reduce energy consumption and enhance reliability during transient events.
Implementation Method 1
a heat transfer coefficient between the air and the server (h)
Implementation Method 2
air mass flow rate (m), and/or temperatures in the data center
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A system and method for evaluating equipment in a data center is disclosed. In one aspect, a method includes receiving parameters for equipment in the data center, the parameters including information descriptive of mass of the equipment, calculating an idealized thermal mass of the equipment based on the received parameters, calculating a temperature associated with the equipment at a first time period of a plurality of time periods based on the idealized thermal mass, and calculating a temperature for each subsequent time period of the plurality of time periods based on the idealized thermal mass and the temperature at a previous time period of the plurality of time periods.