Rack-Level Liquid Flow Control for Direct Injection Servers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current direct-injection liquid-cooled rack information handling systems face challenges in efficiently managing liquid cooling for shared infrastructure servers, particularly in distributing cooling liquids effectively across nodes with varying heat generation rates and form factors within a rack, leading to suboptimal cooling performance and potential overheating.
Innovation Solution
A computer-implemented method and system that utilize a rack-level liquid flow infrastructure controller to dynamically adjust the cooling liquid flow rate for each block of liquid-cooled nodes based on the highest temperature generated, ensuring adequate cooling while optimizing the use of the maximum flow rate cap, and incorporating a modular approach with air-to-liquid heat exchangers for scalable and flexible cooling solutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If cooling liquid is distributed uniformly across all nodes, then cooling system is simple to implement, but nodes with varying heat generation rates experience suboptimal cooling performance and potential overheating
Solution Approach 1:
The patent implements local quality by distributing cooling liquid non-uniformly across nodes based on their individual heat generation characteristics. Each node receives a customized cooling flow rate proportional to its thermal load, ensuring that high-heat nodes receive more cooling while low-heat nodes receive less, thereby optimizing cooling effectiveness for each local condition rather than applying a uniform approach.
2Reliability
If maximum cooling liquid flow rate is allocated to all blocks, then cooling performance is maximized, but system efficiency decreases due to suboptimal use of cooling capacity
Solution Approach 1:
The patent applies dynamics by implementing a dynamic flow rate allocation system that continuously adjusts cooling liquid distribution based on real-time thermal conditions. The system calculates and assigns maximum flow rate caps to each block according to their current heat generation levels, allowing the cooling capacity to adapt dynamically to changing loads rather than maintaining a static maximum flow rate for all blocks.
Solution Approach 2:
The system changes the flow rate parameter dynamically for each block based on thermal conditions. By calculating maximum flow rate caps that reflect actual cooling needs and adjusting allocation accordingly, the system optimizes the flow rate parameter to match demand, preventing both overheating and wasteful over-cooling.
3Ease of manufacture
If cooling system is designed for fixed configuration, then manufacturing and deployment is simplified, but system cannot accommodate different form factors and heat dissipation needs
Solution Approach 1:
The patent implements segmentation by dividing the cooling system into independent block-level units, each capable of autonomous cooling management. This modular architecture allows different form factors and configurations to be accommodated by simply adding or removing blocks, while each block maintains its own cooling control logic. The segmentation enables the system to adapt to varying form factors without requiring complete system redesign.
Solution Approach 2:
The system achieves universality through a standardized block architecture that can accommodate multiple form factors and heat dissipation requirements within a unified cooling framework. Each block is designed with universal interfaces and control mechanisms that allow them to function independently or in combination, enabling the cooling system to serve diverse computational workloads and hardware configurations through a single multi-functional platform.
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 solution enables precise and efficient liquid cooling management, ensuring optimal temperature regulation across nodes with varying heat loads, enhancing the operational integrity of the system by preventing overheating and allowing for flexible configurations to accommodate different form factors and heat dissipation needs.
Implementation Method 1
Each functional component, which generates heat during operation, is in direct contact with the cooling liquid such that the cooling liquid absorbs heat from the functional component
Implementation Method 2
incorporating a modular approach with air-to-liquid heat exchangers for scalable and flexible cooling solutions
Data Source
AI summary
A computer-implemented method controls liquid cooling of a direct injection liquid-cooled (DL) rack information handling system (RIHS). The method includes receiving, at a liquid cooling control subsystem, an incoming cooling liquid supply flow rate corresponding to an incoming cooling liquid supply being supplied to the DL RIHS. A maximum flow rate cap is calculated for each of the LC nodes. The maximum flow rate cap is transmitted to a controller for each of the LC nodes. The controller triggers each of the LC nodes to adjust the associated flow rate for that LC node to correspond to the received maximum flow rate cap for that node.