Understanding the Impact of Data Center Liquid Cooling on Energy and Performance of Machine Learning and Artificial Intelligence Workloads
Journal of Electronic Packaging · 01.06.2025 · 147 (2)
DOI: 10.1115/1.4067136 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Electronic Packaging, online erschienen am 11.12.2024. Verfasst von Bharath Ramakrishnan und 12 weiteren Autor:innen. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 38 Literaturangaben.
Abstract (Originalsprache)
Traditionally, data centers (DC) have used air cooling for IT equipment, but as graphics processing units (GPUs) evolve, they demand more power and sophisticated cooling. Aiming for efficiency, direct liquid cooling (DLC) emerges as a promising solution. We evaluated the effectiveness of DLC versus traditional air cooling on a Microsoft G50 GPU server performing artificial intelligence/machine learning (AI/ML) tasks. The results indicated that DLC greatly enhances GPU performance, increases efficiency by 2.7% in Gflops/s, cuts power usage by 12%, reduces execution times by up to 6.22%, and lowers chip temperatures by 20 °C compared to air cooling. Our research develops an overall performance metric that considers data center, hardware, and chip levels, concluding that DLC is extremely beneficial for AI workloads, increasing energy savings and balancing performance with power requirements.
Zitieren
APA
Ramakrishnan, B., Turner, C., Alissa, H., Trieu, D., Rivera, F., Melton, L., Rao, M., Chigullapalli, S., Getachew, T., Prodanovic, V., Lankston, R., Belady, C., & Oruganti, V. (2025). Understanding the Impact of Data Center Liquid Cooling on Energy and Performance of Machine Learning and Artificial Intelligence Workloads. Journal of Electronic Packaging, 147(2). https://doi.org/10.1115/1.4067136
Vancouver
Ramakrishnan B, Turner C, Alissa H, Trieu D, Rivera F, Melton L, et al. Understanding the Impact of Data Center Liquid Cooling on Energy and Performance of Machine Learning and Artificial Intelligence Workloads. Journal of Electronic Packaging. 2025;147(2). doi:10.1115/1.4067136
BibTeX
@article{Ramakrishnan2025,
author = {Ramakrishnan, Bharath and Turner, Cam and Alissa, Husam and Trieu, Dennis and Rivera, Felipe and Melton, Luke and Rao, Muralikrishna and Chigullapalli, Sruti and Getachew, Tatek and Prodanovic, Vladimir and Lankston, Robert and Belady, Christian and Oruganti, Vaidehi},
title = {Understanding the Impact of Data Center Liquid Cooling on Energy and Performance of Machine Learning and Artificial Intelligence Workloads},
journal = {Journal of Electronic Packaging},
year = {2025},
volume = {147},
number = {2},
doi = {10.1115/1.4067136}
}
EUVIMED