Chillerless Liquid Cooling Control for Data Center Overcooling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data center cooling systems are inefficient in terms of energy consumption, as they often overcool components and struggle to optimize coolant flow based on varying workload and temperature conditions, leading to excessive power usage.
Innovation Solution
A cooling control method that measures temperatures at various points within the system to adjust coolant flow rates and proportions through air-to-liquid heat exchangers, ensuring that only necessary components receive cooling and that coolant flow is optimized based on component and air temperature thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If liquid cooling solutions are implemented to reduce data center cooling power consumption, then energy efficiency is improved, but cooling energy efficiency remains suboptimal due to overcooling and inability to dynamically adjust coolant flow
Solution Approach 1:
The system dynamically adjusts coolant flow rates and temperatures based on real-time component temperatures and workload conditions. Flow rates are modulated according to thermal demands of individual components, allowing the cooling system to adapt its behavior continuously rather than operating at fixed parameters, thereby eliminating overcooling and optimizing energy efficiency.
Solution Approach 2:
Different components receive customized cooling parameters tailored to their specific thermal characteristics and workload demands. The system independently controls coolant flow rates and temperatures for individual components or component groups, allowing each to receive precisely the cooling it needs rather than uniform cooling, thus improving overall cooling energy efficiency.
2Temperature
If coolant flow rate is increased to ensure adequate cooling of all components, then component temperature control is improved, but overall energy consumption increases due to pumping power and excessive cooling of non-critical components
Solution Approach 1:
The system applies differentiated cooling strategies to different components based on their individual thermal requirements and workload conditions. Critical components receiving high computational loads receive higher coolant flow rates and lower temperatures, while non-critical or idle components receive reduced or minimal cooling, thereby maintaining effective temperature control while minimizing pumping energy consumption.
Solution Approach 2:
Instead of providing full cooling capacity to all components uniformly, the system applies partial cooling action only where and when needed. Coolant flow rates are modulated to provide sufficient cooling during high-demand periods and reduced cooling during low-demand periods, eliminating excessive cooling actions that waste pumping energy.
3Device complexity
If uniform coolant flow is provided to all nodes, then system simplicity is maintained, but cooling efficiency decreases as some nodes receive excessive cooling while others may be under-cooled
Solution Approach 1:
The system transitions from uniform coolant flow to localized, component-specific flow control. Each node or component group is equipped with independent flow control mechanisms that adjust coolant distribution based on local thermal conditions and workload demands, thereby achieving optimal cooling efficiency for each location rather than compromising uniformity for simplicity.
Solution Approach 2:
The system dynamically adjusts coolant flow distribution across different nodes based on real-time monitoring of component temperatures and workload conditions. This dynamic adaptation allows the system to optimize cooling efficiency for each node's specific needs while maintaining manageable complexity through centralized control logic that automatically balances performance and simplicity.
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 maximizes cooling efficiency by ensuring that only required components are cooled, reducing overall energy consumption and optimizing power usage in data centers.
Implementation Method 1
an air-to-liquid heat exchanger configured to accept a liquid coolant input and to provide cooled air to the one or more nodes
Implementation Method 2
a liquid cooling system configured to provide liquid coolant to components of the one or more nodes
Data Source
AI summary
Cooling control methods include measuring a temperature of air provided to a plurality of nodes by an air-to-liquid heat exchanger, measuring a temperature of at least one component of the plurality of nodes and finding a maximum component temperature across all such nodes, comparing the maximum component temperature to a first and second component threshold and comparing the air temperature to a first and second air threshold, and controlling a proportion of coolant flow and a coolant flow rate to the air-to-liquid heat exchanger and the plurality of nodes based on the comparisons.