Tschand, A., Rajan, A.T.R., Idgunji, S., Ghosh, A., Holleman, J., Kiraly, C., et al. (2025) MLPerF Power: Benchmarking the Energy Efficiency of Machine Learning Systems from μWatts to MWatts for Sustainable AI. 2025 IEEE International Symposium on High Performance Computer Architecture (HPCA), Las Vegas, 1-5 March 2025, 1201-1216.
https://doi.org/10.1109/hpca61900.2025.00092

has been cited by the following article:

• TITLE: Time Series Models for Predicting Application GPU Utilization and Power Draw

  AUTHORS: Evan Coleman, Dorothy X. Parry, Masha Sosonkina

  KEYWORDS: Time Series Prediction, Performance Modeling, GPU Utilization, Power Draw, Machine Learning, High-Performance Computing

  JOURNAL NAME: Journal of Computer and Communications, Vol.13 No.12, December 19, 2025

  ABSTRACT: This paper explores the application of various time series prediction models to forecast graphical processing unit (GPU) utilization and power draw for machine learning applications using data sets arising from two distinct but representative deep learning inference workloads: the architecturally diverse Inception-v3 and the industry-standard MLPerf ResNet-50. We investigate the use of statistical models, Recurrent Neural Networks (RNNs), and Transformer Neural Networks (TNNs). Our results show that RNNs outperform other models for GPU utilization prediction, while TNNs excel in power draw prediction. These findings may contribute to the development of energy-efficient strategies in high-performance computing environments equipped with GPUs.