TY - GEN
T1 - Measuring the Carbon Intensity of AI in Cloud Instances
AU - Dodge, Jesse
AU - Prewitt, Taylor
AU - Tachet Des Combes, Remi
AU - Odmark, Erika
AU - Schwartz, Roy
AU - Strubell, Emma
AU - Luccioni, Alexandra Sasha
AU - Smith, Noah A.
AU - Decario, Nicole
AU - Buchanan, Will
N1 - Publisher Copyright: © 2022 Owner/Author.
PY - 2022/6/21
Y1 - 2022/6/21
N2 - The advent of cloud computing has provided people around the world with unprecedented access to computational power and enabled rapid growth in technologies such as machine learning, the computational demands of which incur a high energy cost and a commensurate carbon footprint. As a result, recent scholarship has called for better estimates of the greenhouse gas impact of AI: data scientists today do not have easy or reliable access to measurements of this information, which precludes development of actionable tactics. We argue that cloud providers presenting information about software carbon intensity to users is a fundamental stepping stone towards minimizing emissions. In this paper, we provide a framework for measuring software carbon intensity, and propose to measure operational carbon emissions by using location-based and time-specific marginal emissions data per energy unit. We provide measurements of operational software carbon intensity for a set of modern models covering natural language processing and computer vision applications, and a wide range of model sizes, including pretraining of a 6.1 billion parameter language model. We then evaluate a suite of approaches for reducing emissions on the Microsoft Azure cloud compute platform: using cloud instances in different geographic regions, using cloud instances at different times of day, and dynamically pausing cloud instances when the marginal carbon intensity is above a certain threshold. We confirm previous results that the geographic region of the data center plays a significant role in the carbon intensity for a given cloud instance, and find that choosing an appropriate region can have the largest operational emissions reduction impact. We also present new results showing that the time of day has meaningful impact on operational software carbon intensity.Finally, we conclude with recommendations for how machine learning practitioners can use software carbon intensity information to reduce environmental impact.
AB - The advent of cloud computing has provided people around the world with unprecedented access to computational power and enabled rapid growth in technologies such as machine learning, the computational demands of which incur a high energy cost and a commensurate carbon footprint. As a result, recent scholarship has called for better estimates of the greenhouse gas impact of AI: data scientists today do not have easy or reliable access to measurements of this information, which precludes development of actionable tactics. We argue that cloud providers presenting information about software carbon intensity to users is a fundamental stepping stone towards minimizing emissions. In this paper, we provide a framework for measuring software carbon intensity, and propose to measure operational carbon emissions by using location-based and time-specific marginal emissions data per energy unit. We provide measurements of operational software carbon intensity for a set of modern models covering natural language processing and computer vision applications, and a wide range of model sizes, including pretraining of a 6.1 billion parameter language model. We then evaluate a suite of approaches for reducing emissions on the Microsoft Azure cloud compute platform: using cloud instances in different geographic regions, using cloud instances at different times of day, and dynamically pausing cloud instances when the marginal carbon intensity is above a certain threshold. We confirm previous results that the geographic region of the data center plays a significant role in the carbon intensity for a given cloud instance, and find that choosing an appropriate region can have the largest operational emissions reduction impact. We also present new results showing that the time of day has meaningful impact on operational software carbon intensity.Finally, we conclude with recommendations for how machine learning practitioners can use software carbon intensity information to reduce environmental impact.
KW - CO2
KW - carbon awareness
KW - carbon intensity
KW - cloud
KW - emissions
KW - grid
UR - http://www.scopus.com/inward/record.url?scp=85133031151&partnerID=8YFLogxK
U2 - 10.1145/3531146.3533234
DO - 10.1145/3531146.3533234
M3 - منشور من مؤتمر
T3 - ACM International Conference Proceeding Series
SP - 1877
EP - 1894
BT - Proceedings of 2022 5th ACM Conference on Fairness, Accountability, and Transparency, FAccT 2022
T2 - 5th ACM Conference on Fairness, Accountability, and Transparency, FAccT 2022
Y2 - 21 June 2022 through 24 June 2022
ER -
