Temperature Prediction Model for Data Center Layout Variations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional prediction models for temperature management in data centers suffer from insufficient temperature prediction performance due to variations in equipment layout and server installation positions, leading to errors in control assistance.
Innovation Solution
A temperature management system that includes a processor and memory to acquire and process time-series physical quantities, select explanatory variates to minimize errors, generate past case data, and build a local model for predicting future temperatures by associating current input vectors with past case data, thereby improving prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional prediction model is used for temperature management in data centers, then the system can provide basic prediction functionality, but the temperature prediction accuracy deteriorates due to variations in equipment layout and server installation positions
Solution Approach 1:
The system dynamically adapts to different equipment configurations by using historical data from multiple configurations to train the prediction model. The model learns to handle variations in equipment layout and server installation positions by incorporating configuration parameters as input features, allowing it to maintain accuracy across different physical arrangements without requiring manual recalibration for each configuration change
Solution Approach 2:
The prediction model incorporates multiple parameters including equipment configuration data, environmental conditions, and operational metrics. By changing and optimizing these parameters based on actual performance data, the system improves temperature prediction accuracy while maintaining adaptability to different equipment layouts through parameter adjustment rather than structural redesign
2Measurement precision
If more explanatory variates are added to the prediction model to improve accuracy, then the prediction performance improves, but the model complexity increases
Solution Approach 1:
The prediction model is segmented into multiple independent components, each handling specific aspects of temperature prediction. Explanatory variates are grouped into categories such as equipment-related factors, environmental factors, and operational factors. This segmentation allows the system to incorporate comprehensive variables for accurate prediction while managing complexity through modular structure, where each segment can be developed and optimized independently
Solution Approach 2:
The prediction model is designed with universal applicability across different equipment configurations and data center layouts. By creating a multi-functional model that can handle various types of input data (different sensor types, configuration formats, and environmental conditions) through a unified framework, the system achieves high prediction accuracy without proportionally increasing complexity, as the same model structure serves multiple prediction scenarios
3Reliability
If the system stores and processes extensive past case data to improve prediction reliability, then the prediction reliability improves, but the data processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing historical data in optimized formats during off-peak periods. Past case data is pre-aggregated, pre-filtered, and stored with associated metadata that enables rapid retrieval. This preliminary preparation allows the prediction system to quickly access relevant historical cases during operation without performing extensive real-time processing, thus maintaining high reliability while minimizing processing time delays
Solution Approach 2:
The system extracts only the most relevant features and variables from extensive past case data for prediction purposes. Rather than processing entire historical datasets, the model identifies and extracts key explanatory variates and patterns that have the strongest correlation with temperature outcomes. This extraction approach maintains prediction reliability by focusing on critical data elements while significantly reducing the computational burden and processing time required
Data Source
AI summary
An information processing apparatus includes a database configured to store a plurality of physical quantities in time-series, a processor, and a memory storing a program causing the processor to execute acquiring the plurality of physical quantities, selecting first explanatory variates, selecting second explanatory variates, generating past case data by acquiring the physical quantities corresponding to the objective variates and an input variate group of the first explanatory variates and the second explanatory variates, searching for predetermined pieces of past case data in the sequence from the shortest of the inter-vector distances, building up the second model from the input variate group in the predetermined pieces of searched past case data and from the objective variates, and predicting values of objective variates from the second model.