Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study
While the massive adoption of Artificial Intelligence (AI) is threatening the environment, new research efforts begin to be employed to measure and mitigate the carbon footprint of both training and inference phases. In this domain, two carbon-aware training strategies have been proposed in the lite...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-09-01
|
| Series: | Future Internet |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-5903/16/9/334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850261385488891904 |
|---|---|
| author | Roberto Vergallo Luca Mainetti |
| author_facet | Roberto Vergallo Luca Mainetti |
| author_sort | Roberto Vergallo |
| collection | DOAJ |
| description | While the massive adoption of Artificial Intelligence (AI) is threatening the environment, new research efforts begin to be employed to measure and mitigate the carbon footprint of both training and inference phases. In this domain, two carbon-aware training strategies have been proposed in the literature: Flexible Start and Pause & Resume. Such strategies—natively Cloud-based—use the time resource to postpone or pause the training algorithm when the carbon intensity reaches a threshold. While such strategies have proved to achieve interesting results on a benchmark of modern models covering Natural Language Processing (NLP) and computer vision applications and a wide range of model sizes (up to 6.1B parameters), it is still unclear whether such results may hold also with different algorithms and in different geographical regions. In this confirmation study, we use the same methodology as the state-of-the-art strategies to recompute the saving in carbon emissions of Flexible Start and Pause & Resume in the Anomaly Detection (AD) domain. Results confirm their effectiveness in two specific conditions, but the percentage reduction behaves differently compared with what is stated in the existing literature. |
| format | Article |
| id | doaj-art-29084accd66346ccb3eb6511ecbee805 |
| institution | OA Journals |
| issn | 1999-5903 |
| language | English |
| publishDate | 2024-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Future Internet |
| spelling | doaj-art-29084accd66346ccb3eb6511ecbee8052025-08-20T01:55:27ZengMDPI AGFuture Internet1999-59032024-09-0116933410.3390/fi16090334Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation StudyRoberto Vergallo0Luca Mainetti1Department of Innovation Engineering, University of Salento, 73100 Lecce, ItalyDepartment of Innovation Engineering, University of Salento, 73100 Lecce, ItalyWhile the massive adoption of Artificial Intelligence (AI) is threatening the environment, new research efforts begin to be employed to measure and mitigate the carbon footprint of both training and inference phases. In this domain, two carbon-aware training strategies have been proposed in the literature: Flexible Start and Pause & Resume. Such strategies—natively Cloud-based—use the time resource to postpone or pause the training algorithm when the carbon intensity reaches a threshold. While such strategies have proved to achieve interesting results on a benchmark of modern models covering Natural Language Processing (NLP) and computer vision applications and a wide range of model sizes (up to 6.1B parameters), it is still unclear whether such results may hold also with different algorithms and in different geographical regions. In this confirmation study, we use the same methodology as the state-of-the-art strategies to recompute the saving in carbon emissions of Flexible Start and Pause & Resume in the Anomaly Detection (AD) domain. Results confirm their effectiveness in two specific conditions, but the percentage reduction behaves differently compared with what is stated in the existing literature.https://www.mdpi.com/1999-5903/16/9/334green AIgreen softwarecloudsustainabilitycarbon awareness |
| spellingShingle | Roberto Vergallo Luca Mainetti Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study Future Internet green AI green software cloud sustainability carbon awareness |
| title | Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study |
| title_full | Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study |
| title_fullStr | Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study |
| title_full_unstemmed | Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study |
| title_short | Measuring the Effectiveness of Carbon-Aware AI Training Strategies in Cloud Instances: A Confirmation Study |
| title_sort | measuring the effectiveness of carbon aware ai training strategies in cloud instances a confirmation study |
| topic | green AI green software cloud sustainability carbon awareness |
| url | https://www.mdpi.com/1999-5903/16/9/334 |
| work_keys_str_mv | AT robertovergallo measuringtheeffectivenessofcarbonawareaitrainingstrategiesincloudinstancesaconfirmationstudy AT lucamainetti measuringtheeffectivenessofcarbonawareaitrainingstrategiesincloudinstancesaconfirmationstudy |