Data Center Utility Pool Dynamic Power Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern data centers face high energy costs and inefficiencies in cooling systems due to the large heat generated by electronic equipment, which can lead to increased errors and failures in microprocessors, and existing cooling methods are often ineffective in varying ambient conditions.
Innovation Solution
A system that dynamically adjusts the computational demand and cooling capacity by monitoring ambient conditions, such as temperature and humidity, to optimize power usage and allocate resources effectively, using a central control system to manage electrical power distribution and compute load across a data center, allowing for flexible power management and distribution within a common medium voltage power domain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Temperature
If traditional cooling systems are used to remove heat from data centers, then heat removal capability is improved, but energy consumption increases
Solution Approach 1:
The patent implements dynamic power adjustment that continuously monitors ambient conditions (temperature, humidity, wet-bulb temperature) and automatically adjusts data center power consumption accordingly. The system transitions from static cooling capacity to dynamic power management, where compute workload is scaled based on real-time environmental conditions and cooling tower effectiveness, eliminating the need for oversized cooling infrastructure and reducing energy waste during favorable conditions
Solution Approach 2:
The system employs continuous feedback loops that monitor ambient conditions, cooling tower performance, and power consumption. The control system processes wet-bulb temperature data and other environmental parameters to dynamically adjust power allocation, creating a closed-loop control mechanism that optimizes the balance between heat generation and cooling capacity while minimizing energy consumption
2Temperature
If cooling capacity is increased to handle peak heat loads, then heat removal effectiveness is improved, but system complexity increases
Solution Approach 1:
The patent changes the operational parameters of the data center by adjusting power consumption based on ambient conditions rather than maintaining fixed cooling capacity. The system monitors wet-bulb temperature and other parameters to determine optimal power levels, transforming the approach from capacity-based to parameter-based control, which simplifies infrastructure requirements
3Productivity
If power delivery is increased to meet computational demand, then productivity is improved, but heat generation increases
Solution Approach 1:
The system dynamically balances computational output and heat generation by adjusting power delivery according to real-time cooling capacity. When ambient conditions favor heat dissipation, the system increases power delivery to meet computational demand. When conditions deteriorate, power delivery is reduced to match cooling capacity, preventing overheating while maximizing productivity during favorable conditions
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, minimizes the need for energy-intensive cooling components, and maintains optimal operating temperatures, thereby lowering operational costs and improving system reliability by aligning compute tasks with available cooling capacity.
Implementation Method 1
remove heat efficiently from areas that host electronic equipment
Implementation Method 2
the computers generate a large amount of heat
Implementation Method 3
monitoring of ambient conditions indicates that the cooling system will not be able to provide additional cooling
Data Source
AI summary
A system for providing cooled air to electronic equipment includes a cooling system arranged to provide cooling for computer systems in a data center that generate heat from electrical usage; one or more sensors for sensing an environmental condition near the data center; and a central control system programmed to allocate future compute demand for the data center as a function of the sensed environmental condition.