Forecast-Based Data Center Thermal Load Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of thermal management in portable containerized or modular data centers, which are often deployed in unpredictable environments without prior planning, necessitates a real-time optimization of computing loads and cooling strategies to efficiently manage power consumption and operational costs.
Innovation Solution
A method that utilizes forecasts and real-time measurements of environmental conditions to determine optimized computing loads and cooling strategies for networks of containerized or modular data centers, allowing for dynamic distribution of loads and resource management across the data center network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If real-time environmental monitoring and forecasting systems are implemented, then thermal management efficiency is improved, but system complexity and initial costs increase
Solution Approach 1:
The system performs preliminary actions by obtaining forecasts of environmental conditions (temperature, humidity, weather patterns) before they actually occur. This allows the data center to proactively adjust computing loads and cooling strategies in advance, optimizing thermal management before thermal challenges arise, rather than reacting to them in real-time.
Solution Approach 2:
The system dynamically adjusts computing load distribution and cooling resource allocation based on forecasted environmental conditions. The orchestration automatically modifies operational parameters in response to predicted changes in ambient temperature, humidity, and weather patterns, creating a flexible adaptive system that optimizes thermal efficiency continuously.
2Use of energy by moving object
If computing loads are dynamically redistributed across mobile data centers, then power consumption is optimized, but operational complexity increases
Solution Approach 1:
The system implements dynamic load redistribution by continuously monitoring forecasted environmental conditions and automatically adjusting computing load allocation across mobile data centers. The orchestration platform dynamically migrates workloads between data centers based on predicted thermal conditions, optimizing power consumption without manual intervention.
Solution Approach 2:
The system employs feedback mechanisms where operational data from mobile data centers (power consumption, thermal conditions, load levels) is continuously collected and fed back to the orchestration platform. This feedback loop enables automatic adjustment of load distribution strategies based on actual performance and forecasted conditions, optimizing energy efficiency through closed-loop control.
3Adaptability or versatility
If mobile data centers are deployed in unpredictable environments, then deployment flexibility is improved, but thermal management difficulty increases
Solution Approach 1:
The system obtains forecasts of environmental conditions (temperature, humidity, weather patterns) before deploying or operating mobile data centers in new locations. This preliminary information allows operators to anticipate thermal challenges and pre-configurc cooling strategies and load distribution plans, making deployment in unpredictable environments more manageable.
Solution Approach 2:
The orchestration platform acts as an intermediary between the mobile data centers and the unpredictable external environment. It processes forecasted environmental data and translates it into coordinated control actions for load distribution and cooling management, shielding the data centers from the complexity of varying environmental conditions while maintaining deployment flexibility.
Data Source
AI summary
The present disclosure describes techniques evaluating compute and/or thermal loads (among other things) to aid in managing a collection of one or more containerized or modular data centers. For example, forecasts (or real-time measurements) of environmental factors (as well as projected computing demands) may be used to tailor the compute loads, cooling strategies or other metric of data center operations for a network of containerized or modular data centers. Doing so allows an operator of such a data center network to manage specific operational goals in real time.