A Bidirectional Gated Recurrent Unit Model for PUE Prediction in Data Centers

TL;DR

This paper tackles PUE prediction for data centers by integrating a BiGRU-based time-series model with RFECV-driven feature selection using an EnergyPlus-generated Singapore DC dataset of 52,560 samples and 117 features. The authors compare BiGRU to a GRU baseline, finding that BiGRU with 21 selected features yields the best performance (MAE around 3.8e-4, MSE around 5e-7, and $R^2$ near 0.997). The approach combines EnergyPlus data generation, feature selection with XGBoost as the RFECV estimator, and hyperparameter tuning to achieve superior accuracy with reduced input dimensionality. This has practical implications for real-time PUE prediction and energy-efficiency optimization in data centers.

Abstract

Data centers account for significant global energy consumption and a carbon footprint. The recent increasing demand for edge computing and AI advancements drives the growth of data center storage capacity. Energy efficiency is a cost-effective way to combat climate change, cut energy costs, improve business competitiveness, and promote IT and environmental sustainability. Thus, optimizing data center energy management is the most important factor in the sustainability of the world. Power Usage Effectiveness (PUE) is used to represent the operational efficiency of the data center. Predicting PUE using Neural Networks provides an understanding of the effect of each feature on energy consumption, thus enabling targeted modifications of those key features to improve energy efficiency. In this paper, we have developed Bidirectional Gated Recurrent Unit (BiGRU) based PUE prediction model and compared the model performance with GRU. The data set comprises 52,560 samples with 117 features using EnergyPlus, simulating a DC in Singapore. Sets of the most relevant features are selected using the Recursive Feature Elimination with Cross-Validation (RFECV) algorithm for different parameter settings. These feature sets are used to find the optimal hyperparameter configuration and train the BiGRU model. The performance of the optimized BiGRU-based PUE prediction model is then compared with that of GRU using mean squared error (MSE), mean absolute error (MAE), and R-squared metrics.

Figures (5)

• Figure 1: Flowchart of steps involved in the development of PUE prediction model
• Figure 2: EP-Launch setup with Singapore IWEC weather file.
• Figure 3: Number of selected features using RFECV vs. MSE
• Figure 4: BiGRU Hyperparameter tuning for 21 selected features
• Figure 5: BiGRU vs GRU performance over different set of selected features