Towards Energy-Efficient Large-Scale Artificial Intelligence for Sustainable Data Centers
| dc.contributor.author | Dokic, Dusan | |
| dc.contributor.author | Stein, Hannah | |
| dc.contributor.author | Janzen, Sabine | |
| dc.contributor.author | Maaß, Wolfgang | |
| dc.contributor.editor | Klein, Maike | |
| dc.contributor.editor | Krupka, Daniel | |
| dc.contributor.editor | Winter, Cornelia | |
| dc.contributor.editor | Wohlgemuth, Volker | |
| dc.date.accessioned | 2023-11-29T14:50:19Z | |
| dc.date.available | 2023-11-29T14:50:19Z | |
| dc.date.issued | 2023 | |
| dc.description.abstract | The growing interest in AI services has led to a higher demand for computing power to train and execute complex AI models, causing a surge in power consumption in data centers. Together with rising costs for electricity, gas, petroleum, and coal, and the national target for climate neutrality of data centers by 2027, the ability to operate data centers economically is threatened in Germany. To address these issues, a pressing need to improve the sustainability of data centers and that of artificial intelligence. This paper proposes a roadmap to develop sustainable and resource-efficient data centers and AI systems. The roadmap includes four key building blocks: sustainable data centers, AI algorithms, AI sustainability framework, and economic efficiency analysis. Each building block poses pivotal research questions grounded in contemporary literature to guide the pursuit of environmental sustainability in data centers and AI. | en |
| dc.identifier.doi | 10.18420/inf2023_134 | |
| dc.identifier.isbn | 978-3-88579-731-9 | |
| dc.identifier.pissn | 1617-5468 | |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/43057 | |
| dc.language.iso | en | |
| dc.publisher | Gesellschaft für Informatik e.V. | |
| dc.relation.ispartof | INFORMATIK 2023 - Designing Futures: Zukünfte gestalten | |
| dc.relation.ispartofseries | Lecture Notes in Informatics (LNI) - Proceedings, Volume P-337 | |
| dc.subject | Data Centers | |
| dc.subject | Sustainable AI | |
| dc.subject | Energy-Efficient AI Algorithms | |
| dc.title | Towards Energy-Efficient Large-Scale Artificial Intelligence for Sustainable Data Centers | en |
| dc.type | Text/Conference Paper | |
| gi.citation.endPage | 1265 | |
| gi.citation.publisherPlace | Bonn | |
| gi.citation.startPage | 1255 | |
| gi.conference.date | 26.-29. September 2023 | |
| gi.conference.location | Berlin | |
| gi.conference.sessiontitle | Ökologische Nachhaltigkeit - 11. Workshop Umweltinformatik zwischen Nachhaltigkeit und Wandel (UINW 2023) |
Dateien
Originalbündel
1 - 1 von 1