Great Power, Great Responsibility: Recommendations for Reducing Energy for Training Language Models

Joseph McDonald, Baolin Li, Nathan Frey, Devesh Tiwari, Vijay Gadepally, Siddharth Samsi


Abstract
The energy requirements of current natural language processing models continue to grow at a rapid, unsustainable pace. Recent works highlighting this problem conclude there is an urgent need for methods that reduce the energy needs of NLP and machine learning more broadly. In this article, we investigate techniques that can be used to reduce the energy consumption of common NLP applications. In particular, we focus on techniques to measure energy usage and different hardware and datacenter-oriented settings that can be tuned to reduce energy consumption for training and inference for language models. We characterize the impact of these settings on metrics such as computational performance and energy consumption through experiments conducted on a high performance computing system as well as popular cloud computing platforms. These techniques can lead to significant reduction in energy consumption when training language models or their use for inference. For example, power-capping, which limits the maximum power a GPU can consume, can enable a 15% decrease in energy usage with marginal increase in overall computation time when training a transformer-based language model.
Anthology ID:
2022.findings-naacl.151
Volume:
Findings of the Association for Computational Linguistics: NAACL 2022
Month:
July
Year:
2022
Address:
Seattle, United States
Editors:
Marine Carpuat, Marie-Catherine de Marneffe, Ivan Vladimir Meza Ruiz
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1962–1970
Language:
URL:
https://aclanthology.org/2022.findings-naacl.151
DOI:
10.18653/v1/2022.findings-naacl.151
Bibkey:
Cite (ACL):
Joseph McDonald, Baolin Li, Nathan Frey, Devesh Tiwari, Vijay Gadepally, and Siddharth Samsi. 2022. Great Power, Great Responsibility: Recommendations for Reducing Energy for Training Language Models. In Findings of the Association for Computational Linguistics: NAACL 2022, pages 1962–1970, Seattle, United States. Association for Computational Linguistics.
Cite (Informal):
Great Power, Great Responsibility: Recommendations for Reducing Energy for Training Language Models (McDonald et al., Findings 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2022.findings-naacl.151.pdf
Video:
 https://preview.aclanthology.org/nschneid-patch-4/2022.findings-naacl.151.mp4
Data
WikiText-103WikiText-2