Data Center PUE Calculation with Shared Resource Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for calculating Power Usage Effectiveness (PUE) in data centers fail to accurately account for shared resources and energy consumption across multiple data center rooms or facilities, leading to imprecise power usage optimization and inefficiencies.
Innovation Solution
A system and method for calculating PUE that identifies and models shared resource consumption within data centers, adjusting for losses and efficiencies in shared subsystems like cooling and power distribution, allowing for precise measurement and optimization of energy usage across multiple data center rooms and facilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional PUE calculation methods are used, then calculation simplicity is maintained, but measurement precision deteriorates due to failure to account for shared resources and energy consumption across multiple data center rooms
Solution Approach 1:
The patent segments the data center into multiple rooms and identifies shared resources (cooling, power distribution) that serve multiple rooms. By dividing the facility into discrete spatial units and tracking resource allocation at the room level, the system achieves precise PUE calculations for each room while accounting for shared infrastructure consumption.
Solution Approach 2:
The patent introduces a computational model as an intermediary layer between physical infrastructure and measurement. This model represents shared resources and their consumption across rooms, enabling accurate attribution of energy usage without requiring direct physical measurement at every point.
2Loss of energy
If shared resource consumption is not accounted for, then calculation complexity is reduced, but energy waste increases due to imprecise power usage optimization
Solution Approach 1:
The system implements feedback by continuously monitoring shared resource consumption and using this information to optimize power usage. The computational model provides visibility into how shared resources are consumed across rooms, enabling data center operators to make informed decisions that reduce energy waste.
Solution Approach 2:
The computational model serves multiple functions: it tracks power consumption, cooling usage, and other shared resources across multiple rooms simultaneously. This multi-functional approach enables comprehensive energy optimization without proportionally increasing system complexity.
3Measurement precision
If detailed shared resource tracking is implemented, then efficiency metric accuracy is improved, but measurement and calculation complexity increases
Solution Approach 1:
The computational model acts as an intermediary that aggregates and processes measurement data from multiple sources. Rather than requiring direct measurement of every shared resource consumption point, the model synthesizes information from available measurements to derive accurate efficiency metrics for each room.
Solution Approach 2:
The patent creates a virtual copy of the physical data center infrastructure through the computational model. This digital representation mirrors the physical shared resources and their consumption patterns, enabling precise measurement and analysis without the complexity of direct physical measurement at every point.
Data Source
AI summary
A system and method of measuring efficiency of a datacenter is provided. The method includes the acts of identifying at least one space within a model of a datacenter, the at least one space including a first equipment group consuming at least one shared resource provided by at least one shared resource provider, determining an amount of power consumed by the first equipment group, determining an amount of the at least one shared resource consumed by the first equipment group, determining an amount of power consumed by the at least one shared resource, calculating a loss of the first equipment group, and calculating an efficiency metric based on the amount of power consumed by the first equipment group, the amount of power consumed by the at least one shared resource provider, the loss of the first equipment group, and the loss of the at least one shared resource provider.


